Machine Learning Project

( Gender and Emotion Recognition )

Step 1: Import Libraries

#Import Libraries

import os
import sys
import glob
import pickle
import sklearn
import librosa
import warnings
import collections
import numpy as np
import pandas as pd
import seaborn as sns
import librosa.display
import tensorflow as tf
from scipy import stats
from tqdm import notebook
from tensorflow import keras
import IPython.display as ipd
import scipy.io.wavfile as wav
import matplotlib.pyplot as plt
warnings.filterwarnings("ignore")
from sklearn import preprocessing
from prettytable import PrettyTable
from matplotlib.pyplot import figure
tf.config.run_functions_eagerly(True)
from tensorflow.keras.layers import GRU
from astropy.table import Table, Column
from python_speech_features import mfcc
from tensorflow.keras.layers import Dense
from tensorflow.keras.layers import Conv2D
from tensorflow.keras.layers import Conv1D
from tensorflow.keras.layers import Flatten
from tensorflow.keras.models import Sequential
from sklearn.preprocessing import MinMaxScaler
from sklearn.neural_network import MLPClassifier
from tensorflow.keras.utils import to_categorical
from tensorflow.keras.layers import Bidirectional
from tensorflow.keras.layers import TimeDistributed
from imblearn.over_sampling import RandomOverSampler
from sklearn.model_selection import train_test_split
from tensorflow.keras.callbacks import ReduceLROnPlateau, ModelCheckpoint
from sklearn.metrics import accuracy_score, confusion_matrix, balanced_accuracy_score, classification_report
from tensorflow.keras.layers import Dense, Conv1D, MaxPooling1D, Flatten, Dropout, BatchNormalization, LSTM, GlobalAveragePooling1D

Step 2: Load 5 Datasets

Step 2.1: Load Dataset-1

# Load Sample Data

print("\nSample Data:")
print("============\n")
path = ('Datasets/SAVEE Database')
sample_data_1 = [os.path.join(dp, f) for dp, dn, filenames in os.walk(path) 
                 for f in filenames if os.path.splitext(f)[1] == '.wav']

print("\nDisplaying Some Instances from Sample Data: ")
print("===========================================\n")
for i in range(5):
    print("Audio: ")
    ipd.display(ipd.Audio(sample_data_1[i*100]))
    data, sampling_rate = librosa.load(sample_data_1[i*100])
    plt.figure(figsize=(10, 2),facecolor="Green")
    librosa.display.waveplot(data, sampling_rate)

Sample Data:
============


Displaying Some Instances from Sample Data: 
===========================================

Audio: 

Audio: 

Audio: 

Audio: 

Audio: 

Step 2.2: Load Dataset-2

# Load Sample Data

print("\nSample Data:")
print("============\n")
path = ('Datasets/EmoDB')
sample_data_2 = [os.path.join(dp, f) for dp, dn, filenames in os.walk(path) 
                 for f in filenames if os.path.splitext(f)[1] == '.wav']

print("\nDisplaying Some Instances from Sample Data: ")
print("===========================================\n")
for i in range(5):
    print("Audio: ")
    ipd.display(ipd.Audio(sample_data_2[i*100]))
    data, sampling_rate = librosa.load(sample_data_2[i*100])
    plt.figure(figsize=(10, 2),facecolor="blue")
    librosa.display.waveplot(data, sampling_rate)

Sample Data:
============


Displaying Some Instances from Sample Data: 
===========================================

Audio: 

Audio: 

Audio: 

Audio: 

Audio: 

Step 2.3: Load Dataset-3

# Load Sample Data

print("\nSample Data:")
print("============\n")
path = ('Datasets/CREMA-D')
sample_data_3 = [os.path.join(dp, f) for dp, dn, filenames in os.walk(path) 
                 for f in filenames if os.path.splitext(f)[1] == '.wav']

print("\nDisplaying Some Instances from Sample Data: ")
print("===========================================\n")
for i in range(5):
    print("Audio: ")
    ipd.display(ipd.Audio(sample_data_3[i*100]))
    data, sampling_rate = librosa.load(sample_data_3[i*100])
    plt.figure(figsize=(10, 2),facecolor="brown")
    librosa.display.waveplot(data, sampling_rate)

Sample Data:
============


Displaying Some Instances from Sample Data: 
===========================================

Audio: 

Audio: 

Audio: 

Audio: 

Audio: 

Step 2.4: Load Dataset-4

# Load Sample Data

print("\nSample Data:")
print("============\n")
path = ('Datasets/TESS')
sample_data_4 = [os.path.join(dp, f) for dp, dn, filenames in os.walk(path) 
                 for f in filenames if os.path.splitext(f)[1] == '.wav']

print("\nDisplaying Some Instances from Sample Data: ")
print("===========================================\n")
for i in range(5):
    print("Audio: ")
    ipd.display(ipd.Audio(sample_data_4[i*100]))
    data, sampling_rate = librosa.load(sample_data_4[i*100])
    plt.figure(figsize=(10, 2),facecolor="gray")
    librosa.display.waveplot(data, sampling_rate)

Sample Data:
============


Displaying Some Instances from Sample Data: 
===========================================

Audio: 

Audio: 

Audio: 

Audio: 

Audio: 

Step 2.5: Load Dataset-5

# Load Sample Data

print("\nSample Data:")
print("============\n")
path = ('Datasets/RAVDESS')
sample_data_5 = [os.path.join(dp, f) for dp, dn, filenames in os.walk(path) 
                 for f in filenames if os.path.splitext(f)[1] == '.wav']

print("\nDisplaying Some Instances from Sample Data: ")
print("===========================================\n")
for i in range(5):
    print("Audio: ")
    ipd.display(ipd.Audio(sample_data_5[i*100]))
    data, sampling_rate = librosa.load(sample_data_5[i*100])
    plt.figure(figsize=(10, 2),facecolor="yellow")
    librosa.display.waveplot(data, sampling_rate)

Sample Data:
============


Displaying Some Instances from Sample Data: 
===========================================

Audio: 

Audio: 

Audio: 

Audio: 

Audio: 

Step 3: Extract MFCC Features

Step 3.1: Extract MFCC Features from Dataset-1

sample_data_mfcc_features_1 = pd.DataFrame()

for i in notebook.tqdm(range(len(sample_data_1))):
    data, sampling_rate = librosa.load(sample_data_1[i])
    
    # Get MFCC Features
    
    mfcc_features = mfcc(data,sampling_rate,winlen=30,nfft=661500) 
    
    # Convert to DataFrame
    
    mfcc_df = pd.DataFrame(mfcc_features)
    
    if(sample_data_1[i][27:-6] == 'a'):
        mfcc_df["Emotion"]="Anger"
    elif(sample_data_1[i][27:-6] == 'd'):
        mfcc_df["Emotion"]="Disgust"
    elif(sample_data_1[i][27:-6] == 'f'):
        mfcc_df["Emotion"]="Fear"
    elif(sample_data_1[i][27:-6] == 'h'):
        mfcc_df["Emotion"]="Happiness"
    elif(sample_data_1[i][27:-6] == 'n'):
        mfcc_df["Emotion"]="Neutral"
    elif(sample_data_1[i][27:-6] == 'sa'):
        mfcc_df["Emotion"]="Sadness"
    elif(sample_data_1[i][27:-6] == 'su'):
        mfcc_df["Emotion"]="Surprise"
    else:
        mfcc_df["Emotion"]="Other"
    
    mfcc_df["Gender"] = "Male"
    
    sample_data_mfcc_features_1 = sample_data_mfcc_features_1.append(mfcc_df)

# Save to CSV
sample_data_mfcc_features_1.to_csv('Audio Features/sample_data_mfcc_features_1.csv',index= False, mode='w', header=True) 

Step 3.1.1: Print MFCC Features-1

# Load Features
features_1 = pd.read_csv('Audio Features/sample_data_mfcc_features_1.csv')
features_1

	0	1	2	3	4	5	6	7	8	9	10	11	12	Emotion	Gender
0	3.203335	13.795782	-19.908137	23.113571	-41.785224	5.834210	-52.924076	-2.088703	-14.617876	4.936264	-15.795315	11.496223	-4.460716	Anger	Male
1	3.446474	23.732635	-30.598307	4.388828	-28.666846	13.383600	-52.968102	4.408833	2.814173	-10.741523	-14.331456	11.751246	-8.396209	Anger	Male
2	3.476206	17.949951	-30.643702	4.392290	-35.739255	-8.915932	-46.324606	6.564549	-7.061853	-16.553248	-16.938792	8.475676	-20.742700	Anger	Male
3	3.558089	18.276150	-20.346620	-1.583971	-30.147299	0.963494	-54.361284	-9.811369	-10.147195	-11.144327	-1.306936	7.505639	0.572708	Anger	Male
4	3.764346	18.764859	-24.920929	1.265585	-34.814864	-3.281257	-50.514474	-1.195924	-6.026913	-8.751210	-12.911233	8.015905	-6.634611	Anger	Male
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
475	-2.891536	9.977441	-8.494620	20.642597	-43.986634	-5.818822	-42.747704	1.623718	-5.636403	-5.261166	-19.011852	10.283520	-0.037808	Surprise	Male
476	-2.373482	13.664592	-14.080212	22.486546	-38.038503	-3.001456	-38.062197	-3.495917	-15.125975	-11.673487	-15.574197	-0.432495	-6.982044	Surprise	Male
477	-2.244675	21.429362	-11.647796	9.177692	-29.520397	-15.366208	-43.935295	10.374315	-28.431255	-12.655181	-9.804251	-1.886984	4.831945	Surprise	Male
478	-1.600036	13.877167	-13.671807	13.773882	-43.390877	-1.317022	-45.268181	-2.391161	-21.976652	-9.785898	-4.528661	14.398901	-6.945704	Surprise	Male
479	-1.418079	12.886012	-9.240951	11.856341	-56.535884	2.798735	-38.455837	0.368887	-21.125617	-7.183570	-11.136310	10.025685	-2.063207	Surprise	Male

480 rows × 15 columns

Step 3.2: Extract MFCC Features from Dataset-2

sample_data_mfcc_features_2 = pd.DataFrame()

for i in notebook.tqdm(range(len(sample_data_2))):
    data, sampling_rate = librosa.load(sample_data_2[i])
    
    # Get MFCC Features
    
    mfcc_features = mfcc(data,sampling_rate,winlen=30,nfft=661500) 
    
    # Convert to DataFrame
    
    mfcc_df = pd.DataFrame(mfcc_features)
    
    if(sample_data_2[i][24:25] == 'W'):
        mfcc_df["Emotion"]="Anger"
    elif(sample_data_2[i][24:25] == 'E'):
        mfcc_df["Emotion"]="Disgust"
    elif(sample_data_2[i][24:25] == 'A'):
        mfcc_df["Emotion"]="Fear"
    elif(sample_data_2[i][24:25] == 'F'):
        mfcc_df["Emotion"]="Happiness"
    elif(sample_data_2[i][24:25] == 'N'):
        mfcc_df["Emotion"]="Neutral"
    elif(sample_data_2[i][24:25] == 'T'):
        mfcc_df["Emotion"]="Sadness"
    elif(sample_data_2[i][24:25] == 'L'):
        mfcc_df["Emotion"]="Boredom"
    else:
        mfcc_df["Emotion"]="Other"
        
    if(sample_data_2[i][19:21] == '03' or 
       sample_data_2[i][19:21] == '10' or 
       sample_data_2[i][19:21] == '11' or
       sample_data_2[i][19:21] == '12' or
       sample_data_2[i][19:21] == '15' ):
        mfcc_df["Gender"]="Male"
    else:
        mfcc_df["Gender"]="Female"
    
    
    sample_data_mfcc_features_2 = sample_data_mfcc_features_2.append(mfcc_df)

# Save to CSV
sample_data_mfcc_features_2.to_csv('Audio Features/sample_data_mfcc_features_2.csv',index= False, mode='w', header=True) 

Step 3.2.1: Print MFCC Features-2

# Load Features
features_2 = pd.read_csv('Audio Features/sample_data_mfcc_features_2.csv')
features_2

	0	1	2	3	4	5	6	7	8	9	10	11	12	Emotion	Gender
0	2.581027	15.767412	-47.700512	35.579753	-70.270488	46.800920	-58.114879	37.216890	-49.564047	31.066088	-17.529903	3.279682	-12.838690	Happiness	Male
1	2.795378	24.843764	-51.506450	49.454440	-72.968111	55.298546	-74.268759	41.864338	-59.045060	38.514404	0.004737	10.388288	-14.279588	Neutral	Male
2	4.170786	14.397566	-71.067858	40.482368	-84.346462	58.019575	-81.388417	53.770739	-75.670880	40.587622	-35.743369	21.891265	-25.008990	Anger	Male
3	3.337494	16.282372	-50.880033	31.990281	-85.102191	50.261715	-84.640713	46.940764	-82.094347	32.629892	-49.312384	7.776564	-22.946343	Happiness	Male
4	2.707667	24.334550	-45.717779	53.691805	-71.204936	54.309667	-80.084345	44.830838	-61.581425	38.328134	-23.163505	17.898031	-5.970665	Neutral	Male
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
530	3.863628	19.842292	-52.308186	48.280323	-72.358978	54.636782	-87.941838	44.757594	-59.136477	49.136846	-35.821170	15.900585	-32.775497	Boredom	Female
531	4.198406	4.509584	-34.055637	48.250139	-48.483522	48.624353	-72.094906	19.418505	-74.377215	30.325926	-38.960475	-1.243505	-27.763525	Sadness	Female
532	3.450846	1.809250	-26.119918	58.439039	-41.793259	43.821996	-58.959182	33.575929	-56.189126	32.773332	-32.477892	8.302056	-14.578023	Sadness	Female
533	3.457763	7.503398	-88.067662	40.491302	-88.680502	46.283885	-81.724622	27.269826	-67.302466	10.171340	-50.524035	11.235342	-32.179386	Anger	Female
534	4.445711	8.212223	-81.494063	45.600778	-89.101499	32.327346	-92.760920	36.413223	-79.412132	31.038278	-60.340300	26.689882	-26.373804	Anger	Female

535 rows × 15 columns

Step 3.3: Extract MFCC Features from Dataset-3

sample_data_mfcc_features_3 = pd.DataFrame()

female = [1002,1003,1004,1006,1007,1008,1009,1010,1012,1013,1018,1020,1021,1024,1025,1028,1029,1030,1037,1043,1046,1047,1049,
          1052,1053,1054,1055,1056,1058,1060,1061,1063,1072,1073,1074,1075,1076,1078,1079,1082,1084,1089,1091]

for i in notebook.tqdm(range(len(sample_data_3))):
    data, sampling_rate = librosa.load(sample_data_3[i])
    
    # Get MFCC Features
    
    mfcc_features = mfcc(data,sampling_rate,winlen=30,nfft=661500) 
    
    # Convert to DataFrame
    
    mfcc_df = pd.DataFrame(mfcc_features)
    
    if(sample_data_3[i][26:29] == 'ANG'):
        mfcc_df["Emotion"]="Anger"
    elif(sample_data_3[i][26:29] == 'DIS'):
        mfcc_df["Emotion"]="Disgust"
    elif(sample_data_3[i][26:29] == 'FEA'):
        mfcc_df["Emotion"]="Fear"
    elif(sample_data_3[i][26:29] == 'HAP'):
        mfcc_df["Emotion"]="Happiness"
    elif(sample_data_3[i][26:29] == 'NEU'):
        mfcc_df["Emotion"]="Neutral"
    elif(sample_data_3[i][26:29] == 'SAD'):
        mfcc_df["Emotion"]="Sadness"
    else:
        mfcc_df["Emotion"]="Other"
        
    if(int(sample_data_3[i][17:].split('_')[0]) in female):
        mfcc_df["Gender"]="Female"
    else:
        mfcc_df["Gender"]="Male"
    
    
    sample_data_mfcc_features_3 = sample_data_mfcc_features_3.append(mfcc_df)

# Save to CSV
sample_data_mfcc_features_3.to_csv('Audio Features/sample_data_mfcc_features_3.csv',index= False, mode='w', header=True) 

Step 3.3.1: Print MFCC Features-3

# Load Features
features_3 = pd.read_csv('Audio Features/sample_data_mfcc_features_3.csv')
features_3

	0	1	2	3	4	5	6	7	8	9	10	11	12	Emotion	Gender
0	2.368902	23.902528	-65.345819	44.065375	-74.662445	51.996950	-88.516855	53.855236	-78.723548	69.285800	-59.639412	39.177042	-23.609966	Anger	Male
1	0.307114	13.163935	-48.153023	33.389668	-49.739224	34.304775	-69.983706	35.961724	-49.892039	39.105544	-31.849578	27.108298	2.069042	Disgust	Male
2	2.603431	10.167095	-47.088421	13.140881	-51.376008	24.318749	-71.462424	20.749077	-50.124086	37.938860	-31.598722	8.348972	-5.282255	Fear	Male
3	2.140128	11.848147	-47.720734	23.389854	-52.047988	23.094227	-65.452202	24.289967	-50.372034	35.510180	-35.682490	11.530250	8.193065	Happiness	Male
4	0.473232	12.135419	-43.704434	31.896648	-43.032738	21.278558	-59.340704	9.792648	-45.896816	39.344912	-31.064415	19.842668	10.652224	Neutral	Male
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
7437	-0.569248	7.090084	-21.821569	21.218883	-70.987375	58.362293	-53.937357	59.089649	-57.824928	26.901631	-25.225629	5.655281	-13.951756	Disgust	Female
7438	0.275142	-7.106827	-12.841243	33.026974	-53.708565	49.734721	-53.718731	28.586449	-38.315850	26.497463	-24.172343	-5.992458	-15.062148	Fear	Female
7439	0.081269	6.700789	-24.036068	18.961563	-59.892895	46.793140	-65.963296	37.240255	-53.570069	28.635201	-29.254545	-2.764532	-8.967688	Happiness	Female
7440	-0.233033	8.786678	-29.184771	14.561491	-73.676943	47.898016	-35.411657	53.926727	-62.481850	23.092826	-16.729014	1.414838	-16.098669	Neutral	Female
7441	-0.907600	5.432492	-20.481048	23.180579	-67.214288	55.307229	-53.056532	48.972602	-48.267778	34.530610	-17.210164	-2.257076	-11.995452	Sadness	Female

7442 rows × 15 columns

Step 3.4: Extract MFCC Features from Dataset-4

sample_data_mfcc_features_4 = pd.DataFrame()

for i in notebook.tqdm(range(len(sample_data_4))):
    data, sampling_rate = librosa.load(sample_data_4[i])
    
    # Get MFCC Features
    
    mfcc_features = mfcc(data,sampling_rate,winlen=30,nfft=661500) 
    
    # Convert to DataFrame
    
    mfcc_df = pd.DataFrame(mfcc_features)
    
    if(sample_data_4[i].split('_')[2][:-4] == 'angry'):
        mfcc_df["Emotion"]="Anger"
    elif(sample_data_4[i].split('_')[2][:-4] == 'disgust'):
        mfcc_df["Emotion"]="Disgust"
    elif(sample_data_4[i].split('_')[2][:-4] == 'fear'):
        mfcc_df["Emotion"]="Fear"
    elif(sample_data_4[i].split('_')[2][:-4] == 'happy'):
        mfcc_df["Emotion"]="Happiness"
    elif(sample_data_4[i].split('_')[2][:-4] == 'neutral'):
        mfcc_df["Emotion"]="Neutral"
    elif(sample_data_4[i].split('_')[2][:-4] == 'ps'):
        mfcc_df["Emotion"]="Surprise"
    elif(sample_data_4[i].split('_')[2][:-4] == 'sad'):
        mfcc_df["Emotion"]="Sadness"
    else:
        mfcc_df["Emotion"]="Other"
        
    if(sample_data_4[i][14:].split("_")[0] == 'OAF'):
        mfcc_df["Gender"]="Male"
    else:
        mfcc_df["Gender"]="Female"
    
    
    sample_data_mfcc_features_4 = sample_data_mfcc_features_4.append(mfcc_df)

# Save to CSV
sample_data_mfcc_features_4.to_csv('Audio Features/sample_data_mfcc_features_4.csv',index= False, mode='w', header=True) 

Step 3.4.1: Print MFCC Features-4

# Load Features
features_4 = pd.read_csv('Audio Features/sample_data_mfcc_features_4.csv')
features_4

	0	1	2	3	4	5	6	7	8	9	10	11	12	Emotion	Gender
0	1.220228	-6.288021	-23.058824	-42.426226	-32.405139	-2.121633	-32.409469	-19.628282	-12.538443	-8.152727	8.038784	9.805581	-8.588909	Anger	Male
1	-0.977203	-4.524827	-8.214732	-29.952360	-8.692126	-8.750833	-22.322727	-14.788581	-4.259685	-10.710336	-10.794590	-6.866066	-7.855454	Disgust	Male
2	0.054579	-3.640480	-8.949627	-38.282103	-26.989143	-16.127692	-31.411542	-17.400726	-23.480958	-4.422475	-20.796203	-9.274272	-12.435063	Fear	Male
3	-0.666131	-4.405894	-1.385126	-29.719358	-15.141410	0.250221	-31.969870	-16.251581	-9.591181	-11.897906	-3.236711	-10.320902	-0.489421	Happiness	Male
4	-2.182739	-4.278151	-1.558268	-21.337291	-21.234993	-10.585688	-17.260342	-23.917076	-15.028336	-1.602557	-13.839702	-1.707235	-11.335032	Neutral	Male
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
2795	1.889912	-9.972459	-19.383733	-16.514340	3.237600	-40.164680	-21.195187	-36.141423	-33.244664	-11.505766	-13.660636	4.744601	-3.159517	Fear	Female
2796	3.691365	-16.734092	-30.044690	-8.298291	-18.335224	-40.469660	-37.680667	-54.758465	0.051854	-16.442842	-20.674230	6.932490	-33.305323	Happiness	Female
2797	0.734736	-11.863582	-6.351481	-8.360497	-4.854154	-32.655983	-13.512118	-33.856513	-16.572519	-2.119042	-19.188326	-4.721083	-16.395590	Neutral	Female
2798	0.830298	-14.009006	-14.546528	-22.484116	-20.190596	-34.711502	-20.611750	-41.547508	-24.357789	-9.737252	-4.848874	-0.929159	-17.349877	Surprise	Female
2799	1.380444	-14.983023	12.742822	-0.159556	-5.243630	-32.397391	-11.955386	-27.468668	-13.824717	-9.303697	-8.919692	1.806700	-19.386223	Sadness	Female

2800 rows × 15 columns

Step 3.5: Extract MFCC Features from Dataset-5

sample_data_mfcc_features_5 = pd.DataFrame()

for i in notebook.tqdm(range(len(sample_data_5))):
    data, sampling_rate = librosa.load(sample_data_5[i])
    
    # Get MFCC Features
    
    mfcc_features = mfcc(data,sampling_rate,winlen=30,nfft=661500) 
    
    # Convert to DataFrame
    
    mfcc_df = pd.DataFrame(mfcc_features)
    
    if(sample_data_5[i][26:].split('-')[2] == '05'):
        mfcc_df["Emotion"]="Anger"
    elif(sample_data_5[i][26:].split('-')[2] == '07'):
        mfcc_df["Emotion"]="Disgust"
    elif(sample_data_5[i][26:].split('-')[2] == '06'):
        mfcc_df["Emotion"]="Fear"
    elif(sample_data_5[i][26:].split('-')[2] == '03'):
        mfcc_df["Emotion"]="Happiness"
    elif(sample_data_5[i][26:].split('-')[2] == '01'):
        mfcc_df["Emotion"]="Neutral"
    elif(sample_data_5[i][26:].split('-')[2] == '08'):
        mfcc_df["Emotion"]="Surprise"
    elif(sample_data_5[i][26:].split('-')[2] == '04'):
        mfcc_df["Emotion"]="Sadness"
    else:
        mfcc_df["Emotion"]="Other"
        
    if(int(sample_data_5[i][26:].split('-')[6][:-4])%2 == 1):
        mfcc_df["Gender"]="Male"
    else:
        mfcc_df["Gender"]="Female"
    
    
    sample_data_mfcc_features_5 = sample_data_mfcc_features_5.append(mfcc_df)

# Save to CSV
sample_data_mfcc_features_5.to_csv('Audio Features/sample_data_mfcc_features_5.csv',index= False, mode='w', header=True) 

Step 3.5.1: Print MFCC Features-5

# Load Features
features_5 = pd.read_csv('Audio Features/sample_data_mfcc_features_5.csv')
features_5

	0	1	2	3	4	5	6	7	8	9	10	11	12	Emotion	Gender
0	-3.359473	-2.146878	-11.564124	1.891171	-9.008204	-10.801787	-10.833871	-15.028368	-8.398560	-9.163281	10.721436	4.608740	1.473425	Neutral	Male
1	-3.329815	-2.676908	-7.403491	0.345950	-9.717465	-12.311797	-5.109952	-14.550771	-11.636229	-8.670857	10.890765	3.436975	2.016244	Neutral	Male
2	-2.444367	-7.687208	-3.431780	0.146838	-24.630812	-7.976574	-3.713535	-15.636945	-3.625836	-14.900788	8.883719	8.981573	6.187388	Neutral	Male
3	-2.511114	-8.663157	-2.275137	-2.383363	-22.228988	-12.477076	0.356150	-11.706877	-1.475497	-11.278868	8.459507	14.609624	1.657197	Neutral	Male
4	-3.891415	-5.680255	-3.246147	-0.504579	-6.481897	-0.268197	-19.338007	-15.519278	-4.290236	-7.981147	7.386760	1.352189	0.599136	Other	Male
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1435	-0.020836	-16.578694	-20.212497	-35.196251	-41.204122	-11.407926	-28.910504	-26.953338	-28.814963	-18.164941	-22.130170	-20.616013	-21.943507	Surprise	Female
1436	0.895214	-14.870234	-24.218205	-33.666994	-30.253341	-31.812652	-12.578500	-27.321653	-18.336286	-10.580679	-31.383302	-8.684033	-16.206074	Surprise	Female
1437	0.696340	-15.306912	-23.957822	-27.642590	-27.104280	-34.365583	-12.110144	-30.969206	-18.769509	-5.478475	-22.061271	-1.286473	-12.172666	Surprise	Female
1438	2.295254	-27.182203	-12.448132	-28.649016	-35.397557	-14.519178	-28.478788	-27.172286	-24.487222	-20.706953	-13.494822	-9.909802	-19.705105	Surprise	Female
1439	1.574269	-20.212374	-5.617051	-38.268733	-36.260999	-8.539351	-21.161402	-15.748326	-30.735166	-22.535320	-13.359412	-3.140330	-1.425395	Surprise	Female

1440 rows × 15 columns

Step 4: Pre-Processing

Step 4.1: Combine Feature of all Datasets

# Name of all 5 Datasets
frames = [features_1, features_2, features_3, features_4, features_5]

# Combined Dataset 
combined = pd.concat(frames)
combined = combined.reset_index(drop=True)

# Save to CSV
combined.to_csv('Audio Features/combined_features.csv',index= False, mode='w', header=True)
combined

	0	1	2	3	4	5	6	7	8	9	10	11	12	Emotion	Gender
0	3.203335	13.795782	-19.908137	23.113571	-41.785224	5.834210	-52.924076	-2.088703	-14.617876	4.936264	-15.795315	11.496223	-4.460716	Anger	Male
1	3.446474	23.732635	-30.598307	4.388828	-28.666846	13.383600	-52.968102	4.408833	2.814173	-10.741523	-14.331456	11.751246	-8.396209	Anger	Male
2	3.476206	17.949951	-30.643702	4.392290	-35.739255	-8.915932	-46.324606	6.564549	-7.061853	-16.553248	-16.938792	8.475676	-20.742700	Anger	Male
3	3.558089	18.276150	-20.346620	-1.583971	-30.147299	0.963494	-54.361284	-9.811369	-10.147195	-11.144327	-1.306936	7.505639	0.572708	Anger	Male
4	3.764346	18.764859	-24.920929	1.265585	-34.814864	-3.281257	-50.514474	-1.195924	-6.026913	-8.751210	-12.911233	8.015905	-6.634611	Anger	Male
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
12692	-0.020836	-16.578694	-20.212497	-35.196251	-41.204122	-11.407926	-28.910504	-26.953338	-28.814963	-18.164941	-22.130170	-20.616013	-21.943507	Surprise	Female
12693	0.895214	-14.870234	-24.218205	-33.666994	-30.253341	-31.812652	-12.578500	-27.321653	-18.336286	-10.580679	-31.383302	-8.684033	-16.206074	Surprise	Female
12694	0.696340	-15.306912	-23.957822	-27.642590	-27.104280	-34.365583	-12.110144	-30.969206	-18.769509	-5.478475	-22.061271	-1.286473	-12.172666	Surprise	Female
12695	2.295254	-27.182203	-12.448132	-28.649016	-35.397557	-14.519178	-28.478788	-27.172286	-24.487222	-20.706953	-13.494822	-9.909802	-19.705105	Surprise	Female
12696	1.574269	-20.212374	-5.617051	-38.268733	-36.260999	-8.539351	-21.161402	-15.748326	-30.735166	-22.535320	-13.359412	-3.140330	-1.425395	Surprise	Female

12697 rows × 15 columns

Step 4.2: Check Data Distribution

# Plotting

plt.style.use('fivethirtyeight')
combined=pd.read_csv('Audio Features/combined_features.csv')
df_filtered = combined[combined['Emotion'] != "Boredom"]
combined = df_filtered[df_filtered['Emotion'] != "Other"]

figure(figsize=(14, 6), dpi=80)
plt.title('Count of Emotions', size=25)
sns.countplot(combined['Emotion'])
plt.ylabel('Count', size=14)
plt.xlabel('Emotions', size=14)
sns.despine(top=True, right=True, left=False, bottom=False)
plt.show()

# Plotting

figure(figsize=(7, 6), dpi=80)
plt.title('Count of Gender', size=25)
sns.countplot(combined['Gender'])
plt.ylabel('Count', size=14)
plt.xlabel('Genders', size=14)
sns.despine(top=True, right=True, left=False, bottom=False)
plt.show()

Step 4.3: Data Sampling

# Random Over Sampling

X = combined.drop(['Emotion'], axis = 1)
y = combined['Emotion']

ros = RandomOverSampler(random_state=64)
X_resampled, y_resampled = ros.fit_resample(X, y)

# Finding Frequency of each Class before and After Random Over Sampling

counter_1 = collections.Counter(y)
counter_2 = collections.Counter(y_resampled)

# Before Sampling

plt.figure()
fig = plt.gcf()
fig.set_size_inches(24, 5)
plt.subplot(1, 2, 1)
plt.bar(counter_1.keys(), counter_1.values())
plt.xticks([0,1,2,3,4,5,6])
plt.title("Before Sampling")

# After Sampling

plt.figure()
fig = plt.gcf()
fig.set_size_inches(24, 5)
plt.subplot(1, 2, 2)
plt.bar(counter_2.keys(), counter_2.values(), color = 'green')
plt.xticks([0,1,2,3,4,5,6])
plt.title("After Sampling")

plt.subplots_adjust(top=0.92, bottom=0.08, left=0.10, right=0.95, hspace=0.5,wspace=0.35)
plt.show()

Step 4.4: Data Scaling

# Scaling

y1 = y_resampled
y2 = X_resampled["Gender"]

X = X_resampled.drop(["Gender"],axis=1)

scaler = MinMaxScaler()
X_scaled = scaler.fit_transform(X)

print(y1.shape)
print(y2.shape)
print(X_scaled.shape)

print(X_scaled)

(14350,)
(14350,)
(14350, 13)
[[0.93610433 0.65309955 0.62192149 ... 0.60099951 0.52473204 0.56745201]
 [0.94190358 0.7893224  0.53968236 ... 0.61542485 0.52713997 0.51623315]
 [0.94261276 0.71004844 0.53933314 ... 0.58973133 0.49621191 0.35554854]
 ...
 [0.90833302 0.15584516 0.68150021 ... 0.7242391  0.34603792 0.46732237]
 [0.89425745 0.28430183 0.62634095 ... 0.56568075 0.33727729 0.38896452]
 [0.83385165 0.40645843 0.78249741 ... 0.69382561 0.39112332 0.49386858]]

Step 4.5: Label Encoding Outputs

# Emotions
encoder_1 = preprocessing.LabelEncoder()
y1_encoded = encoder_1.fit_transform(y1)

# Gender
encoder_2 = preprocessing.LabelEncoder()
y2_encoded = encoder_2.fit_transform(y2)

Step 5: Split Dataset in Traing and Testing

labels=[]
i=0
while i<len(y1_encoded):
    label=[]
    label.append([y1_encoded[i]])
    label.append([y2_encoded[i]])
    labels.append(label)
    i+=1
labels = np.array(labels)

X_train, X_test, Y_train, Y_test= train_test_split(X_scaled, labels, test_size=0.1, random_state = 64)

Y_train = [Y_train[:,1],Y_train[:,0]]
Y_test = [Y_test[:,1],Y_test[:,0]]

X_train = X_train
X_test = X_test
y1_train = Y_train[0]
y1_test = Y_test[0]
y2_train = Y_train[1]
y2_test = Y_test[1]

print(X_train.shape)
print(X_test.shape)
print(y1_train.shape)
print(y1_test.shape)
print(y2_train.shape)
print(y2_test.shape)

(12915, 13)
(1435, 13)
(12915, 1)
(1435, 1)
(12915, 1)
(1435, 1)

df = pd.DataFrame(y1_train)
df["1"] = y2_train
y_train = np.array(df)
print(y_train)

[[0 6]
 [1 4]
 [1 0]
 ...
 [1 4]
 [1 3]
 [1 1]]

df = pd.DataFrame(y1_test)
df["1"] = y2_test
y_test = np.array(df)
print(y_test)

[[0 1]
 [0 4]
 [1 6]
 ...
 [1 4]
 [0 6]
 [0 4]]

Step 6: Train CNN Classifier

Step 6.1: For Gender Classification

x_train, x_test, y_train, y_test = X_train, X_test, y1_train, y1_test

size = 13
x_train = np.array(x_train)
y_train = np.array(y_train)
x_test = np.array(x_test) 
y_test = np.array(y_test)
x_train = x_train.reshape(x_train.shape[0], size, 1)
x_test = x_test.reshape(x_test.shape[0], size, 1)
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)

# Create Model

model_1_g=Sequential()
model_1_g.add(Conv1D(512, kernel_size=5, strides=1, padding='same', activation='relu', input_shape=(x_train.shape[1], 1)))
model_1_g.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))

model_1_g.add(Conv1D(256, kernel_size=5, strides=1, padding='same', activation='relu'))
model_1_g.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))

model_1_g.add(Conv1D(128, kernel_size=5, strides=1, padding='same', activation='relu'))
model_1_g.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))
model_1_g.add(Dropout(0.2))

model_1_g.add(Conv1D(64, kernel_size=5, strides=1, padding='same', activation='relu'))
model_1_g.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))

model_1_g.add(GlobalAveragePooling1D())

model_1_g.add(Flatten())
model_1_g.add(Dense(units=32, activation='relu'))
model_1_g.add(Dropout(0.3))

model_1_g.add(Dense(units=2, activation='softmax'))
model_1_g.compile(optimizer = 'adam' , loss = 'categorical_crossentropy' , metrics = ['accuracy'])

model_1_g.summary()

Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv1d (Conv1D)              (None, 13, 512)           3072      
_________________________________________________________________
max_pooling1d (MaxPooling1D) (None, 7, 512)            0         
_________________________________________________________________
conv1d_1 (Conv1D)            (None, 7, 256)            655616    
_________________________________________________________________
max_pooling1d_1 (MaxPooling1 (None, 4, 256)            0         
_________________________________________________________________
conv1d_2 (Conv1D)            (None, 4, 128)            163968    
_________________________________________________________________
max_pooling1d_2 (MaxPooling1 (None, 2, 128)            0         
_________________________________________________________________
dropout (Dropout)            (None, 2, 128)            0         
_________________________________________________________________
conv1d_3 (Conv1D)            (None, 2, 64)             41024     
_________________________________________________________________
max_pooling1d_3 (MaxPooling1 (None, 1, 64)             0         
_________________________________________________________________
global_average_pooling1d (Gl (None, 64)                0         
_________________________________________________________________
flatten (Flatten)            (None, 64)                0         
_________________________________________________________________
dense (Dense)                (None, 32)                2080      
_________________________________________________________________
dropout_1 (Dropout)          (None, 32)                0         
_________________________________________________________________
dense_1 (Dense)              (None, 2)                 66        
=================================================================
Total params: 865,826
Trainable params: 865,826
Non-trainable params: 0
_________________________________________________________________

Step 6.1.1: Training

# Training

rlrp = ReduceLROnPlateau(monitor='loss', factor=0.4, verbose=0, patience=2, min_lr=0.0000001)
history_1_g = model_1_g.fit(x_train, y_train, batch_size=64, epochs=250, validation_data=(x_test, y_test), callbacks=[rlrp])

Epoch 1/250
202/202 [==============================] - 42s 207ms/step - loss: 0.5645 - accuracy: 0.6802 - val_loss: 0.5028 - val_accuracy: 0.7185
Epoch 2/250
202/202 [==============================] - 43s 213ms/step - loss: 0.4373 - accuracy: 0.7782 - val_loss: 0.4200 - val_accuracy: 0.7735
Epoch 3/250
202/202 [==============================] - 48s 236ms/step - loss: 0.3995 - accuracy: 0.8050 - val_loss: 0.3989 - val_accuracy: 0.8028
Epoch 4/250
202/202 [==============================] - 55s 272ms/step - loss: 0.3805 - accuracy: 0.8184 - val_loss: 0.3773 - val_accuracy: 0.8188
Epoch 5/250
202/202 [==============================] - 54s 269ms/step - loss: 0.3583 - accuracy: 0.8274 - val_loss: 0.3656 - val_accuracy: 0.8237
Epoch 6/250
202/202 [==============================] - 65s 323ms/step - loss: 0.3460 - accuracy: 0.8355 - val_loss: 0.3502 - val_accuracy: 0.8272
Epoch 7/250
202/202 [==============================] - 58s 286ms/step - loss: 0.3328 - accuracy: 0.8455 - val_loss: 0.3486 - val_accuracy: 0.8383
Epoch 8/250
202/202 [==============================] - 53s 265ms/step - loss: 0.3277 - accuracy: 0.8458 - val_loss: 0.3520 - val_accuracy: 0.8362
Epoch 9/250
202/202 [==============================] - 42s 208ms/step - loss: 0.3101 - accuracy: 0.8601 - val_loss: 0.3322 - val_accuracy: 0.8362
Epoch 10/250
202/202 [==============================] - 38s 188ms/step - loss: 0.3088 - accuracy: 0.8598 - val_loss: 0.3044 - val_accuracy: 0.8627
Epoch 11/250
202/202 [==============================] - 39s 192ms/step - loss: 0.3034 - accuracy: 0.8612 - val_loss: 0.3046 - val_accuracy: 0.8551
Epoch 12/250
202/202 [==============================] - 37s 185ms/step - loss: 0.2932 - accuracy: 0.8697 - val_loss: 0.3013 - val_accuracy: 0.8544
Epoch 13/250
202/202 [==============================] - 47s 232ms/step - loss: 0.2923 - accuracy: 0.8708 - val_loss: 0.2989 - val_accuracy: 0.8592
Epoch 14/250
202/202 [==============================] - 45s 223ms/step - loss: 0.2794 - accuracy: 0.8740 - val_loss: 0.2998 - val_accuracy: 0.8578
Epoch 15/250
202/202 [==============================] - 52s 258ms/step - loss: 0.2713 - accuracy: 0.8777 - val_loss: 0.3005 - val_accuracy: 0.8746
Epoch 16/250
202/202 [==============================] - 69s 339ms/step - loss: 0.2688 - accuracy: 0.8791 - val_loss: 0.2858 - val_accuracy: 0.8808
Epoch 17/250
202/202 [==============================] - 41s 203ms/step - loss: 0.2594 - accuracy: 0.8860 - val_loss: 0.2897 - val_accuracy: 0.8704
Epoch 18/250
202/202 [==============================] - 39s 195ms/step - loss: 0.2561 - accuracy: 0.8869 - val_loss: 0.2715 - val_accuracy: 0.8801
Epoch 19/250
202/202 [==============================] - 40s 199ms/step - loss: 0.2578 - accuracy: 0.8848 - val_loss: 0.2958 - val_accuracy: 0.8599
Epoch 20/250
202/202 [==============================] - 40s 199ms/step - loss: 0.2440 - accuracy: 0.8924 - val_loss: 0.3527 - val_accuracy: 0.8000
Epoch 21/250
202/202 [==============================] - 40s 197ms/step - loss: 0.2404 - accuracy: 0.8948 - val_loss: 0.3081 - val_accuracy: 0.8662
Epoch 22/250
202/202 [==============================] - 50s 249ms/step - loss: 0.2357 - accuracy: 0.8945 - val_loss: 0.2574 - val_accuracy: 0.8885
Epoch 23/250
202/202 [==============================] - 47s 233ms/step - loss: 0.2418 - accuracy: 0.8945 - val_loss: 0.2533 - val_accuracy: 0.8878
Epoch 24/250
202/202 [==============================] - 48s 237ms/step - loss: 0.2305 - accuracy: 0.8964 - val_loss: 0.2608 - val_accuracy: 0.8892
Epoch 25/250
202/202 [==============================] - 47s 234ms/step - loss: 0.2266 - accuracy: 0.9002 - val_loss: 0.3107 - val_accuracy: 0.8530
Epoch 26/250
202/202 [==============================] - 48s 239ms/step - loss: 0.2278 - accuracy: 0.8966 - val_loss: 0.2671 - val_accuracy: 0.8836
Epoch 27/250
202/202 [==============================] - 52s 257ms/step - loss: 0.2164 - accuracy: 0.9092 - val_loss: 0.2744 - val_accuracy: 0.8718
Epoch 28/250
202/202 [==============================] - 55s 272ms/step - loss: 0.2197 - accuracy: 0.9020 - val_loss: 0.2615 - val_accuracy: 0.8760
Epoch 29/250
202/202 [==============================] - 50s 247ms/step - loss: 0.2118 - accuracy: 0.9068 - val_loss: 0.2535 - val_accuracy: 0.8899
Epoch 30/250
202/202 [==============================] - 51s 253ms/step - loss: 0.2094 - accuracy: 0.9111 - val_loss: 0.2456 - val_accuracy: 0.8976
Epoch 31/250
202/202 [==============================] - 41s 204ms/step - loss: 0.2007 - accuracy: 0.9132 - val_loss: 0.2331 - val_accuracy: 0.8955
Epoch 32/250
202/202 [==============================] - 41s 205ms/step - loss: 0.2003 - accuracy: 0.9131 - val_loss: 0.2396 - val_accuracy: 0.9024
Epoch 33/250
202/202 [==============================] - 48s 237ms/step - loss: 0.2024 - accuracy: 0.9105 - val_loss: 0.2479 - val_accuracy: 0.8899
Epoch 34/250
202/202 [==============================] - 49s 244ms/step - loss: 0.1929 - accuracy: 0.9172 - val_loss: 0.2347 - val_accuracy: 0.9017
Epoch 35/250
202/202 [==============================] - 47s 233ms/step - loss: 0.1954 - accuracy: 0.9162 - val_loss: 0.2692 - val_accuracy: 0.8843
Epoch 36/250
202/202 [==============================] - 43s 215ms/step - loss: 0.1864 - accuracy: 0.9168 - val_loss: 0.2473 - val_accuracy: 0.8990
Epoch 37/250
202/202 [==============================] - 45s 222ms/step - loss: 0.1906 - accuracy: 0.9181 - val_loss: 0.2510 - val_accuracy: 0.8955
Epoch 38/250
202/202 [==============================] - 40s 198ms/step - loss: 0.1835 - accuracy: 0.9212 - val_loss: 0.2407 - val_accuracy: 0.9045
Epoch 39/250
202/202 [==============================] - 42s 206ms/step - loss: 0.1844 - accuracy: 0.9202 - val_loss: 0.2316 - val_accuracy: 0.9059
Epoch 40/250
202/202 [==============================] - 39s 193ms/step - loss: 0.1841 - accuracy: 0.9216 - val_loss: 0.2383 - val_accuracy: 0.9031
Epoch 41/250
202/202 [==============================] - 47s 231ms/step - loss: 0.1486 - accuracy: 0.9387 - val_loss: 0.2286 - val_accuracy: 0.9157
Epoch 42/250
202/202 [==============================] - 41s 201ms/step - loss: 0.1316 - accuracy: 0.9441 - val_loss: 0.2316 - val_accuracy: 0.9199
Epoch 43/250
202/202 [==============================] - 48s 236ms/step - loss: 0.1325 - accuracy: 0.9449 - val_loss: 0.2303 - val_accuracy: 0.9199
Epoch 44/250
202/202 [==============================] - 47s 231ms/step - loss: 0.1344 - accuracy: 0.9416 - val_loss: 0.2564 - val_accuracy: 0.9108
Epoch 45/250
202/202 [==============================] - 39s 194ms/step - loss: 0.1119 - accuracy: 0.9522 - val_loss: 0.2376 - val_accuracy: 0.9213
Epoch 46/250
202/202 [==============================] - 39s 195ms/step - loss: 0.1099 - accuracy: 0.9520 - val_loss: 0.2383 - val_accuracy: 0.9268
Epoch 47/250
202/202 [==============================] - 43s 211ms/step - loss: 0.1049 - accuracy: 0.9539 - val_loss: 0.2725 - val_accuracy: 0.9143
Epoch 48/250
202/202 [==============================] - 43s 214ms/step - loss: 0.1052 - accuracy: 0.9566 - val_loss: 0.2437 - val_accuracy: 0.9261
Epoch 49/250
202/202 [==============================] - 44s 218ms/step - loss: 0.1031 - accuracy: 0.9552 - val_loss: 0.2460 - val_accuracy: 0.9213
Epoch 50/250
202/202 [==============================] - 51s 251ms/step - loss: 0.0967 - accuracy: 0.9587 - val_loss: 0.2561 - val_accuracy: 0.9261
Epoch 51/250
202/202 [==============================] - 45s 221ms/step - loss: 0.1003 - accuracy: 0.9577 - val_loss: 0.2555 - val_accuracy: 0.9254
Epoch 52/250
202/202 [==============================] - 42s 207ms/step - loss: 0.0959 - accuracy: 0.9593 - val_loss: 0.2413 - val_accuracy: 0.9261
Epoch 53/250
202/202 [==============================] - 38s 188ms/step - loss: 0.0931 - accuracy: 0.9617 - val_loss: 0.2544 - val_accuracy: 0.9185
Epoch 54/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0935 - accuracy: 0.9597 - val_loss: 0.2617 - val_accuracy: 0.9157
Epoch 55/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0894 - accuracy: 0.9617 - val_loss: 0.2859 - val_accuracy: 0.9150
Epoch 56/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0923 - accuracy: 0.9617 - val_loss: 0.2652 - val_accuracy: 0.9275
Epoch 57/250
202/202 [==============================] - 40s 196ms/step - loss: 0.0887 - accuracy: 0.9617 - val_loss: 0.2605 - val_accuracy: 0.9206
Epoch 58/250
202/202 [==============================] - 44s 219ms/step - loss: 0.0874 - accuracy: 0.9629 - val_loss: 0.2667 - val_accuracy: 0.9254
Epoch 59/250
202/202 [==============================] - 40s 198ms/step - loss: 0.0842 - accuracy: 0.9636 - val_loss: 0.2639 - val_accuracy: 0.9296
Epoch 60/250
202/202 [==============================] - 39s 194ms/step - loss: 0.0865 - accuracy: 0.9650 - val_loss: 0.2629 - val_accuracy: 0.9303
Epoch 61/250
202/202 [==============================] - 39s 192ms/step - loss: 0.0869 - accuracy: 0.9638 - val_loss: 0.2552 - val_accuracy: 0.9247
Epoch 62/250
202/202 [==============================] - 39s 192ms/step - loss: 0.0750 - accuracy: 0.9695 - val_loss: 0.2659 - val_accuracy: 0.9289
Epoch 63/250
202/202 [==============================] - 40s 200ms/step - loss: 0.0710 - accuracy: 0.9707 - val_loss: 0.2690 - val_accuracy: 0.9261
Epoch 64/250
202/202 [==============================] - 42s 208ms/step - loss: 0.0725 - accuracy: 0.9709 - val_loss: 0.2793 - val_accuracy: 0.9282
Epoch 65/250
202/202 [==============================] - 41s 202ms/step - loss: 0.0695 - accuracy: 0.9714 - val_loss: 0.2776 - val_accuracy: 0.9226
Epoch 66/250
202/202 [==============================] - 41s 203ms/step - loss: 0.0706 - accuracy: 0.9708 - val_loss: 0.2794 - val_accuracy: 0.9275
Epoch 67/250
202/202 [==============================] - 41s 201ms/step - loss: 0.0669 - accuracy: 0.9732 - val_loss: 0.2830 - val_accuracy: 0.9226
Epoch 68/250
202/202 [==============================] - 40s 197ms/step - loss: 0.0687 - accuracy: 0.9707 - val_loss: 0.2848 - val_accuracy: 0.9247
Epoch 69/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0659 - accuracy: 0.9734 - val_loss: 0.2941 - val_accuracy: 0.9261
Epoch 70/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0666 - accuracy: 0.9724 - val_loss: 0.2886 - val_accuracy: 0.9352
Epoch 71/250
202/202 [==============================] - 37s 185ms/step - loss: 0.0668 - accuracy: 0.9724 - val_loss: 0.2995 - val_accuracy: 0.9289
Epoch 72/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0614 - accuracy: 0.9752 - val_loss: 0.2957 - val_accuracy: 0.9352
Epoch 73/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0624 - accuracy: 0.9744 - val_loss: 0.2919 - val_accuracy: 0.9296
Epoch 74/250
202/202 [==============================] - 38s 189ms/step - loss: 0.0585 - accuracy: 0.9773 - val_loss: 0.2968 - val_accuracy: 0.9338
Epoch 75/250
202/202 [==============================] - 38s 189ms/step - loss: 0.0587 - accuracy: 0.9752 - val_loss: 0.3011 - val_accuracy: 0.9317
Epoch 76/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0614 - accuracy: 0.9759 - val_loss: 0.2992 - val_accuracy: 0.9345
Epoch 77/250
202/202 [==============================] - 38s 190ms/step - loss: 0.0590 - accuracy: 0.9772 - val_loss: 0.2977 - val_accuracy: 0.9324
Epoch 78/250
202/202 [==============================] - 38s 189ms/step - loss: 0.0591 - accuracy: 0.9754 - val_loss: 0.3000 - val_accuracy: 0.9317
Epoch 79/250
202/202 [==============================] - 38s 190ms/step - loss: 0.0563 - accuracy: 0.9771 - val_loss: 0.3013 - val_accuracy: 0.9324
Epoch 80/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0592 - accuracy: 0.9774 - val_loss: 0.3010 - val_accuracy: 0.9338
Epoch 81/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0590 - accuracy: 0.9755 - val_loss: 0.3013 - val_accuracy: 0.9345
Epoch 82/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0559 - accuracy: 0.9765 - val_loss: 0.3013 - val_accuracy: 0.9331
Epoch 83/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0547 - accuracy: 0.9776 - val_loss: 0.3017 - val_accuracy: 0.9331
Epoch 84/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0572 - accuracy: 0.9770 - val_loss: 0.3023 - val_accuracy: 0.9338
Epoch 85/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0568 - accuracy: 0.9774 - val_loss: 0.3024 - val_accuracy: 0.9338
Epoch 86/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0561 - accuracy: 0.9781 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 87/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0568 - accuracy: 0.9768 - val_loss: 0.3021 - val_accuracy: 0.9331
Epoch 88/250
202/202 [==============================] - 39s 191ms/step - loss: 0.0567 - accuracy: 0.9788 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 89/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0569 - accuracy: 0.9782 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 90/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0546 - accuracy: 0.9779 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 91/250
202/202 [==============================] - 46s 228ms/step - loss: 0.0556 - accuracy: 0.9784 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 92/250
202/202 [==============================] - 42s 209ms/step - loss: 0.0583 - accuracy: 0.9775 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 93/250
202/202 [==============================] - 40s 197ms/step - loss: 0.0578 - accuracy: 0.9767 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 94/250
202/202 [==============================] - 38s 189ms/step - loss: 0.0575 - accuracy: 0.9770 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 95/250
202/202 [==============================] - 48s 238ms/step - loss: 0.0578 - accuracy: 0.9775 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 96/250
202/202 [==============================] - 51s 251ms/step - loss: 0.0558 - accuracy: 0.9781 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 97/250
202/202 [==============================] - 46s 226ms/step - loss: 0.0556 - accuracy: 0.9786 - val_loss: 0.3022 - val_accuracy: 0.9331
Epoch 98/250
202/202 [==============================] - 51s 254ms/step - loss: 0.0567 - accuracy: 0.9781 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 99/250
202/202 [==============================] - 46s 227ms/step - loss: 0.0552 - accuracy: 0.9782 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 100/250
202/202 [==============================] - 39s 192ms/step - loss: 0.0571 - accuracy: 0.9767 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 101/250
202/202 [==============================] - 38s 187ms/step - loss: 0.0580 - accuracy: 0.9773 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 102/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0559 - accuracy: 0.9779 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 103/250
202/202 [==============================] - 44s 218ms/step - loss: 0.0558 - accuracy: 0.9771 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 104/250
202/202 [==============================] - 48s 238ms/step - loss: 0.0552 - accuracy: 0.9784 - val_loss: 0.3024 - val_accuracy: 0.9324
Epoch 105/250
202/202 [==============================] - 46s 225ms/step - loss: 0.0558 - accuracy: 0.9782 - val_loss: 0.3024 - val_accuracy: 0.9324
Epoch 106/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0568 - accuracy: 0.9776 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 107/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0575 - accuracy: 0.9776 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 108/250
202/202 [==============================] - 37s 185ms/step - loss: 0.0557 - accuracy: 0.9791 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 109/250
202/202 [==============================] - 40s 196ms/step - loss: 0.0564 - accuracy: 0.9781 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 110/250
202/202 [==============================] - 38s 187ms/step - loss: 0.0564 - accuracy: 0.9782 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 111/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0562 - accuracy: 0.9779 - val_loss: 0.3023 - val_accuracy: 0.9331
Epoch 112/250
202/202 [==============================] - 42s 206ms/step - loss: 0.0558 - accuracy: 0.9781 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 113/250
202/202 [==============================] - 42s 206ms/step - loss: 0.0555 - accuracy: 0.9780 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 114/250
202/202 [==============================] - 43s 215ms/step - loss: 0.0584 - accuracy: 0.9775 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 115/250
202/202 [==============================] - 44s 218ms/step - loss: 0.0561 - accuracy: 0.9789 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 116/250
202/202 [==============================] - 49s 243ms/step - loss: 0.0564 - accuracy: 0.9764 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 117/250
202/202 [==============================] - 44s 220ms/step - loss: 0.0570 - accuracy: 0.9776 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 118/250
202/202 [==============================] - 50s 246ms/step - loss: 0.0575 - accuracy: 0.9769 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 119/250
202/202 [==============================] - 42s 210ms/step - loss: 0.0572 - accuracy: 0.9773 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 120/250
202/202 [==============================] - 45s 222ms/step - loss: 0.0572 - accuracy: 0.9779 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 121/250
202/202 [==============================] - 42s 206ms/step - loss: 0.0565 - accuracy: 0.9782 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 122/250
202/202 [==============================] - 51s 255ms/step - loss: 0.0563 - accuracy: 0.9775 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 123/250
202/202 [==============================] - 51s 254ms/step - loss: 0.0550 - accuracy: 0.9792 - val_loss: 0.3024 - val_accuracy: 0.9331
Epoch 124/250
202/202 [==============================] - 64s 314ms/step - loss: 0.0569 - accuracy: 0.9774 - val_loss: 0.3024 - val_accuracy: 0.9324
Epoch 125/250
202/202 [==============================] - 52s 256ms/step - loss: 0.0558 - accuracy: 0.9779 - val_loss: 0.3024 - val_accuracy: 0.9324
Epoch 126/250
202/202 [==============================] - 44s 220ms/step - loss: 0.0564 - accuracy: 0.9770 - val_loss: 0.3025 - val_accuracy: 0.9324
Epoch 127/250
202/202 [==============================] - 43s 215ms/step - loss: 0.0568 - accuracy: 0.9776 - val_loss: 0.3025 - val_accuracy: 0.9331
Epoch 128/250
202/202 [==============================] - 39s 194ms/step - loss: 0.0581 - accuracy: 0.9767 - val_loss: 0.3025 - val_accuracy: 0.9331
Epoch 129/250
202/202 [==============================] - 39s 192ms/step - loss: 0.0575 - accuracy: 0.9771 - val_loss: 0.3025 - val_accuracy: 0.9331
Epoch 130/250
202/202 [==============================] - 40s 197ms/step - loss: 0.0560 - accuracy: 0.9778 - val_loss: 0.3025 - val_accuracy: 0.9331
Epoch 131/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0561 - accuracy: 0.9776 - val_loss: 0.3025 - val_accuracy: 0.9331
Epoch 132/250
202/202 [==============================] - 38s 189ms/step - loss: 0.0548 - accuracy: 0.9786 - val_loss: 0.3025 - val_accuracy: 0.9331
Epoch 133/250
202/202 [==============================] - 38s 188ms/step - loss: 0.0571 - accuracy: 0.9776 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 134/250
202/202 [==============================] - 38s 187ms/step - loss: 0.0566 - accuracy: 0.9783 - val_loss: 0.3026 - val_accuracy: 0.9324
Epoch 135/250
202/202 [==============================] - 42s 207ms/step - loss: 0.0553 - accuracy: 0.9774 - val_loss: 0.3026 - val_accuracy: 0.9324
Epoch 136/250
202/202 [==============================] - 37s 185ms/step - loss: 0.0561 - accuracy: 0.9783 - val_loss: 0.3026 - val_accuracy: 0.9324
Epoch 137/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0557 - accuracy: 0.9788 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 138/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0567 - accuracy: 0.9781 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 139/250
202/202 [==============================] - 39s 192ms/step - loss: 0.0559 - accuracy: 0.9776 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 140/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0556 - accuracy: 0.9775 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 141/250
202/202 [==============================] - 36s 180ms/step - loss: 0.0549 - accuracy: 0.9779 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 142/250
202/202 [==============================] - 42s 208ms/step - loss: 0.0551 - accuracy: 0.9779 - val_loss: 0.3027 - val_accuracy: 0.9331
Epoch 143/250
202/202 [==============================] - 38s 187ms/step - loss: 0.0571 - accuracy: 0.9768 - val_loss: 0.3026 - val_accuracy: 0.9331
Epoch 144/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0559 - accuracy: 0.9776 - val_loss: 0.3027 - val_accuracy: 0.9331
Epoch 145/250
202/202 [==============================] - 36s 180ms/step - loss: 0.0549 - accuracy: 0.9787 - val_loss: 0.3027 - val_accuracy: 0.9331
Epoch 146/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0557 - accuracy: 0.9779 - val_loss: 0.3027 - val_accuracy: 0.9324
Epoch 147/250
202/202 [==============================] - 36s 180ms/step - loss: 0.0567 - accuracy: 0.9779 - val_loss: 0.3027 - val_accuracy: 0.9324
Epoch 148/250
202/202 [==============================] - 36s 180ms/step - loss: 0.0579 - accuracy: 0.9772 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 149/250
202/202 [==============================] - 37s 185ms/step - loss: 0.0558 - accuracy: 0.9772 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 150/250
202/202 [==============================] - 44s 217ms/step - loss: 0.0562 - accuracy: 0.9778 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 151/250
202/202 [==============================] - 41s 202ms/step - loss: 0.0561 - accuracy: 0.9777 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 152/250
202/202 [==============================] - 40s 200ms/step - loss: 0.0564 - accuracy: 0.9765 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 153/250
202/202 [==============================] - 40s 198ms/step - loss: 0.0568 - accuracy: 0.9769 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 154/250
202/202 [==============================] - 41s 203ms/step - loss: 0.0564 - accuracy: 0.9781 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 155/250
202/202 [==============================] - 40s 197ms/step - loss: 0.0565 - accuracy: 0.9779 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 156/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0568 - accuracy: 0.9768 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 157/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0560 - accuracy: 0.9776 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 158/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0555 - accuracy: 0.9773 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 159/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0571 - accuracy: 0.9772 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 160/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0562 - accuracy: 0.9772 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 161/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0556 - accuracy: 0.9772 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 162/250
202/202 [==============================] - 37s 185ms/step - loss: 0.0556 - accuracy: 0.9783 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 163/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0554 - accuracy: 0.9779 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 164/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0573 - accuracy: 0.9761 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 165/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0560 - accuracy: 0.9768 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 166/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0559 - accuracy: 0.9779 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 167/250
202/202 [==============================] - 35s 172ms/step - loss: 0.0559 - accuracy: 0.9769 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 168/250
202/202 [==============================] - 34s 170ms/step - loss: 0.0551 - accuracy: 0.9779 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 169/250
202/202 [==============================] - 34s 168ms/step - loss: 0.0559 - accuracy: 0.9772 - val_loss: 0.3028 - val_accuracy: 0.9324
Epoch 170/250
202/202 [==============================] - 34s 168ms/step - loss: 0.0566 - accuracy: 0.9783 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 171/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0568 - accuracy: 0.9765 - val_loss: 0.3028 - val_accuracy: 0.9331
Epoch 172/250
202/202 [==============================] - 34s 170ms/step - loss: 0.0554 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 173/250
202/202 [==============================] - 34s 168ms/step - loss: 0.0569 - accuracy: 0.9772 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 174/250
202/202 [==============================] - 34s 168ms/step - loss: 0.0566 - accuracy: 0.9779 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 175/250
202/202 [==============================] - 35s 171ms/step - loss: 0.0557 - accuracy: 0.9769 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 176/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0561 - accuracy: 0.9783 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 177/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0551 - accuracy: 0.9782 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 178/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0577 - accuracy: 0.9769 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 179/250
202/202 [==============================] - 34s 169ms/step - loss: 0.0560 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 180/250
202/202 [==============================] - 34s 169ms/step - loss: 0.0562 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9324
Epoch 181/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0555 - accuracy: 0.9774 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 182/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0551 - accuracy: 0.9782 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 183/250
202/202 [==============================] - 34s 169ms/step - loss: 0.0564 - accuracy: 0.9783 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 184/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0553 - accuracy: 0.9773 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 185/250
202/202 [==============================] - 34s 168ms/step - loss: 0.0584 - accuracy: 0.9767 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 186/250
202/202 [==============================] - 34s 169ms/step - loss: 0.0570 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 187/250
202/202 [==============================] - 34s 169ms/step - loss: 0.0567 - accuracy: 0.9772 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 188/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0564 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 189/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0567 - accuracy: 0.9776 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 190/250
202/202 [==============================] - 34s 169ms/step - loss: 0.0577 - accuracy: 0.9770 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 191/250
202/202 [==============================] - 34s 167ms/step - loss: 0.0588 - accuracy: 0.9761 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 192/250
202/202 [==============================] - 36s 179ms/step - loss: 0.0559 - accuracy: 0.9779 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 193/250
202/202 [==============================] - 48s 236ms/step - loss: 0.0566 - accuracy: 0.9776 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 194/250
202/202 [==============================] - 49s 243ms/step - loss: 0.0552 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 195/250
202/202 [==============================] - 48s 238ms/step - loss: 0.0560 - accuracy: 0.9777 - val_loss: 0.3030 - val_accuracy: 0.9324
Epoch 196/250
202/202 [==============================] - 46s 230ms/step - loss: 0.0546 - accuracy: 0.9783 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 197/250
202/202 [==============================] - 46s 229ms/step - loss: 0.0581 - accuracy: 0.9767 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 198/250
202/202 [==============================] - 47s 233ms/step - loss: 0.0552 - accuracy: 0.9784 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 199/250
202/202 [==============================] - 47s 230ms/step - loss: 0.0564 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 200/250
202/202 [==============================] - 48s 235ms/step - loss: 0.0554 - accuracy: 0.9774 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 201/250
202/202 [==============================] - 48s 237ms/step - loss: 0.0566 - accuracy: 0.9775 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 202/250
202/202 [==============================] - 48s 237ms/step - loss: 0.0553 - accuracy: 0.9772 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 203/250
202/202 [==============================] - 48s 237ms/step - loss: 0.0556 - accuracy: 0.9775 - val_loss: 0.3030 - val_accuracy: 0.9331
Epoch 204/250
202/202 [==============================] - 47s 235ms/step - loss: 0.0558 - accuracy: 0.9784 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 205/250
202/202 [==============================] - 48s 237ms/step - loss: 0.0557 - accuracy: 0.9783 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 206/250
202/202 [==============================] - 49s 244ms/step - loss: 0.0546 - accuracy: 0.9770 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 207/250
202/202 [==============================] - 46s 226ms/step - loss: 0.0564 - accuracy: 0.9772 - val_loss: 0.3029 - val_accuracy: 0.9331
Epoch 208/250
202/202 [==============================] - 46s 228ms/step - loss: 0.0550 - accuracy: 0.9782 - val_loss: 0.3030 - val_accuracy: 0.9331
Epoch 209/250
202/202 [==============================] - 46s 228ms/step - loss: 0.0555 - accuracy: 0.9775 - val_loss: 0.3030 - val_accuracy: 0.9331
Epoch 210/250
202/202 [==============================] - 46s 227ms/step - loss: 0.0570 - accuracy: 0.9772 - val_loss: 0.3030 - val_accuracy: 0.9324
Epoch 211/250
202/202 [==============================] - 47s 231ms/step - loss: 0.0545 - accuracy: 0.9784 - val_loss: 0.3030 - val_accuracy: 0.9324
Epoch 212/250
202/202 [==============================] - 46s 230ms/step - loss: 0.0552 - accuracy: 0.9775 - val_loss: 0.3030 - val_accuracy: 0.9331
Epoch 213/250
202/202 [==============================] - 46s 230ms/step - loss: 0.0575 - accuracy: 0.9775 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 214/250
202/202 [==============================] - 47s 232ms/step - loss: 0.0569 - accuracy: 0.9779 - val_loss: 0.3031 - val_accuracy: 0.9324
Epoch 215/250
202/202 [==============================] - 47s 230ms/step - loss: 0.0558 - accuracy: 0.9778 - val_loss: 0.3031 - val_accuracy: 0.9324
Epoch 216/250
202/202 [==============================] - 47s 232ms/step - loss: 0.0572 - accuracy: 0.9773 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 217/250
202/202 [==============================] - 47s 230ms/step - loss: 0.0557 - accuracy: 0.9781 - val_loss: 0.3030 - val_accuracy: 0.9331
Epoch 218/250
202/202 [==============================] - 46s 229ms/step - loss: 0.0566 - accuracy: 0.9774 - val_loss: 0.3030 - val_accuracy: 0.9331
Epoch 219/250
202/202 [==============================] - 47s 234ms/step - loss: 0.0567 - accuracy: 0.9767 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 220/250
202/202 [==============================] - 48s 239ms/step - loss: 0.0568 - accuracy: 0.9771 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 221/250
202/202 [==============================] - 51s 251ms/step - loss: 0.0545 - accuracy: 0.9781 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 222/250
202/202 [==============================] - 56s 278ms/step - loss: 0.0562 - accuracy: 0.9784 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 223/250
202/202 [==============================] - 56s 278ms/step - loss: 0.0562 - accuracy: 0.9783 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 224/250
202/202 [==============================] - 56s 278ms/step - loss: 0.0545 - accuracy: 0.9799 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 225/250
202/202 [==============================] - 52s 258ms/step - loss: 0.0555 - accuracy: 0.9775 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 226/250
202/202 [==============================] - 48s 239ms/step - loss: 0.0565 - accuracy: 0.9779 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 227/250
202/202 [==============================] - 49s 243ms/step - loss: 0.0543 - accuracy: 0.9779 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 228/250
202/202 [==============================] - 54s 268ms/step - loss: 0.0557 - accuracy: 0.9784 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 229/250
202/202 [==============================] - 46s 228ms/step - loss: 0.0563 - accuracy: 0.9776 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 230/250
202/202 [==============================] - 51s 252ms/step - loss: 0.0565 - accuracy: 0.9760 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 231/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0570 - accuracy: 0.9771 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 232/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0567 - accuracy: 0.9768 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 233/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0555 - accuracy: 0.9780 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 234/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0573 - accuracy: 0.9764 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 235/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0563 - accuracy: 0.9775 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 236/250
202/202 [==============================] - 37s 184ms/step - loss: 0.0557 - accuracy: 0.9772 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 237/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0555 - accuracy: 0.9773 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 238/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0577 - accuracy: 0.9765 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 239/250
202/202 [==============================] - 37s 185ms/step - loss: 0.0572 - accuracy: 0.9768 - val_loss: 0.3031 - val_accuracy: 0.9331
Epoch 240/250
202/202 [==============================] - 37s 181ms/step - loss: 0.0553 - accuracy: 0.9781 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 241/250
202/202 [==============================] - 37s 181ms/step - loss: 0.0569 - accuracy: 0.9768 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 242/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0551 - accuracy: 0.9778 - val_loss: 0.3032 - val_accuracy: 0.9331
Epoch 243/250
202/202 [==============================] - 37s 181ms/step - loss: 0.0560 - accuracy: 0.9779 - val_loss: 0.3033 - val_accuracy: 0.9324
Epoch 244/250
202/202 [==============================] - 37s 181ms/step - loss: 0.0562 - accuracy: 0.9772 - val_loss: 0.3033 - val_accuracy: 0.9324
Epoch 245/250
202/202 [==============================] - 37s 181ms/step - loss: 0.0553 - accuracy: 0.9786 - val_loss: 0.3033 - val_accuracy: 0.9324
Epoch 246/250
202/202 [==============================] - 38s 186ms/step - loss: 0.0574 - accuracy: 0.9783 - val_loss: 0.3033 - val_accuracy: 0.9331
Epoch 247/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0565 - accuracy: 0.9779 - val_loss: 0.3033 - val_accuracy: 0.9331
Epoch 248/250
202/202 [==============================] - 37s 183ms/step - loss: 0.0571 - accuracy: 0.9765 - val_loss: 0.3033 - val_accuracy: 0.9331
Epoch 249/250
202/202 [==============================] - 38s 187ms/step - loss: 0.0575 - accuracy: 0.9769 - val_loss: 0.3034 - val_accuracy: 0.9324
Epoch 250/250
202/202 [==============================] - 37s 182ms/step - loss: 0.0572 - accuracy: 0.9778 - val_loss: 0.3034 - val_accuracy: 0.9324

Step 6.1.2: Learning Curve

plt.style.use('fivethirtyeight')
epochs = [i for i in range(250)]
fig , ax = plt.subplots(1,2)
train_acc = history_1_g.history['accuracy']
train_loss = history_1_g.history['loss']
test_acc = history_1_g.history['val_accuracy']
test_loss = history_1_g.history['val_loss']

fig.set_size_inches(20,6)
ax[0].plot(epochs , train_loss , label = 'Training Loss')
ax[0].plot(epochs , test_loss , label = 'Testing Loss')
ax[0].set_title('Training & Testing Loss')
ax[0].legend()
ax[0].set_xlabel("Epochs")

ax[1].plot(epochs , train_acc , label = 'Training Accuracy')
ax[1].plot(epochs , test_acc , label = 'Testing Accuracy')
ax[1].set_title('Training & Testing Accuracy')
ax[1].legend()
ax[1].set_xlabel("Epochs")
plt.show()

Step 6.1.3: Save model

# Save Model

model_1_g.save('Trained Models/model_1_g.h5')

Step 6.2: For Emotion Classification

x_train, x_test, y_train, y_test = X_train, X_test, y2_train, y2_test

size = 13
x_train = np.array(x_train)
y_train = np.array(y_train)
x_test = np.array(x_test) 
y_test = np.array(y_test)
x_train = x_train.reshape(x_train.shape[0], size, 1)
x_test = x_test.reshape(x_test.shape[0], size, 1)
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)

# Create Model

model_1_e=Sequential()
model_1_e.add(Conv1D(512, kernel_size=5, strides=1, padding='same', activation='relu', input_shape=(x_train.shape[1], 1)))
model_1_e.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))

model_1_e.add(Conv1D(256, kernel_size=5, strides=1, padding='same', activation='relu'))
model_1_e.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))

model_1_e.add(Conv1D(128, kernel_size=5, strides=1, padding='same', activation='relu'))
model_1_e.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))
model_1_e.add(Dropout(0.2))

model_1_e.add(Conv1D(64, kernel_size=5, strides=1, padding='same', activation='relu'))
model_1_e.add(MaxPooling1D(pool_size=5, strides = 2, padding = 'same'))

model_1_e.add(GlobalAveragePooling1D())

model_1_e.add(Flatten())
model_1_e.add(Dense(units=32, activation='relu'))
model_1_e.add(Dropout(0.3))

model_1_e.add(Dense(units=7, activation='softmax'))
model_1_e.compile(optimizer = 'adam' , loss = 'categorical_crossentropy' , metrics = ['accuracy'])

model_1_e.summary()

Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv1d_4 (Conv1D)            (None, 13, 512)           3072      
_________________________________________________________________
max_pooling1d_4 (MaxPooling1 (None, 7, 512)            0         
_________________________________________________________________
conv1d_5 (Conv1D)            (None, 7, 256)            655616    
_________________________________________________________________
max_pooling1d_5 (MaxPooling1 (None, 4, 256)            0         
_________________________________________________________________
conv1d_6 (Conv1D)            (None, 4, 128)            163968    
_________________________________________________________________
max_pooling1d_6 (MaxPooling1 (None, 2, 128)            0         
_________________________________________________________________
dropout_2 (Dropout)          (None, 2, 128)            0         
_________________________________________________________________
conv1d_7 (Conv1D)            (None, 2, 64)             41024     
_________________________________________________________________
max_pooling1d_7 (MaxPooling1 (None, 1, 64)             0         
_________________________________________________________________
global_average_pooling1d_1 ( (None, 64)                0         
_________________________________________________________________
flatten_1 (Flatten)          (None, 64)                0         
_________________________________________________________________
dense_2 (Dense)              (None, 32)                2080      
_________________________________________________________________
dropout_3 (Dropout)          (None, 32)                0         
_________________________________________________________________
dense_3 (Dense)              (None, 7)                 231       
=================================================================
Total params: 865,991
Trainable params: 865,991
Non-trainable params: 0
_________________________________________________________________

Step 6.2.1: Training

# Training

rlrp = ReduceLROnPlateau(monitor='loss', factor=0.4, verbose=0, patience=2, min_lr=0.0000001)
history_1_e = model_1_e.fit(x_train, y_train, batch_size=64, epochs=250, validation_data=(x_test, y_test), callbacks=[rlrp])

Epoch 1/250
202/202 [==============================] - 37s 185ms/step - loss: 1.8635 - accuracy: 0.2314 - val_loss: 1.6696 - val_accuracy: 0.3477
Epoch 2/250
202/202 [==============================] - 38s 187ms/step - loss: 1.6155 - accuracy: 0.3628 - val_loss: 1.4881 - val_accuracy: 0.4202
Epoch 3/250
202/202 [==============================] - 37s 185ms/step - loss: 1.5331 - accuracy: 0.3982 - val_loss: 1.4356 - val_accuracy: 0.4432
Epoch 4/250
202/202 [==============================] - 38s 188ms/step - loss: 1.4914 - accuracy: 0.4230 - val_loss: 1.4228 - val_accuracy: 0.4481
Epoch 5/250
202/202 [==============================] - 37s 182ms/step - loss: 1.4481 - accuracy: 0.4389 - val_loss: 1.3948 - val_accuracy: 0.4585
Epoch 6/250
202/202 [==============================] - 34s 168ms/step - loss: 1.3961 - accuracy: 0.4591 - val_loss: 1.3350 - val_accuracy: 0.4829
Epoch 7/250
202/202 [==============================] - 34s 168ms/step - loss: 1.3646 - accuracy: 0.4735 - val_loss: 1.3087 - val_accuracy: 0.4843
Epoch 8/250
202/202 [==============================] - 34s 168ms/step - loss: 1.3466 - accuracy: 0.4791 - val_loss: 1.3082 - val_accuracy: 0.4892
Epoch 9/250
202/202 [==============================] - 34s 171ms/step - loss: 1.3302 - accuracy: 0.4873 - val_loss: 1.2862 - val_accuracy: 0.4892
Epoch 10/250
202/202 [==============================] - 34s 168ms/step - loss: 1.2999 - accuracy: 0.4965 - val_loss: 1.2905 - val_accuracy: 0.4941
Epoch 11/250
202/202 [==============================] - 34s 169ms/step - loss: 1.2871 - accuracy: 0.5010 - val_loss: 1.2598 - val_accuracy: 0.5087
Epoch 12/250
202/202 [==============================] - 34s 170ms/step - loss: 1.2646 - accuracy: 0.5110 - val_loss: 1.2192 - val_accuracy: 0.5143
Epoch 13/250
202/202 [==============================] - 34s 168ms/step - loss: 1.2385 - accuracy: 0.5175 - val_loss: 1.2148 - val_accuracy: 0.5164
Epoch 14/250
202/202 [==============================] - 35s 171ms/step - loss: 1.2303 - accuracy: 0.5240 - val_loss: 1.2316 - val_accuracy: 0.5178
Epoch 15/250
202/202 [==============================] - 34s 168ms/step - loss: 1.2150 - accuracy: 0.5254 - val_loss: 1.1838 - val_accuracy: 0.5261
Epoch 16/250
202/202 [==============================] - 34s 170ms/step - loss: 1.2007 - accuracy: 0.5337 - val_loss: 1.1699 - val_accuracy: 0.5261
Epoch 17/250
202/202 [==============================] - 34s 169ms/step - loss: 1.1976 - accuracy: 0.5340 - val_loss: 1.1837 - val_accuracy: 0.5394
Epoch 18/250
202/202 [==============================] - 34s 168ms/step - loss: 1.1913 - accuracy: 0.5367 - val_loss: 1.1741 - val_accuracy: 0.5366
Epoch 19/250
202/202 [==============================] - 34s 168ms/step - loss: 1.1591 - accuracy: 0.5510 - val_loss: 1.1787 - val_accuracy: 0.5380
Epoch 20/250
202/202 [==============================] - 34s 170ms/step - loss: 1.1728 - accuracy: 0.5471 - val_loss: 1.1441 - val_accuracy: 0.5512
Epoch 21/250
202/202 [==============================] - 34s 168ms/step - loss: 1.1389 - accuracy: 0.5597 - val_loss: 1.1294 - val_accuracy: 0.5645
Epoch 22/250
202/202 [==============================] - 35s 171ms/step - loss: 1.1412 - accuracy: 0.5555 - val_loss: 1.1447 - val_accuracy: 0.5603
Epoch 23/250
202/202 [==============================] - 34s 170ms/step - loss: 1.1297 - accuracy: 0.5695 - val_loss: 1.1319 - val_accuracy: 0.5679
Epoch 24/250
202/202 [==============================] - 34s 170ms/step - loss: 1.1219 - accuracy: 0.5686 - val_loss: 1.1366 - val_accuracy: 0.5540
Epoch 25/250
202/202 [==============================] - 34s 168ms/step - loss: 1.1180 - accuracy: 0.5659 - val_loss: 1.1410 - val_accuracy: 0.5617
Epoch 26/250
202/202 [==============================] - 34s 170ms/step - loss: 1.1202 - accuracy: 0.5683 - val_loss: 1.1301 - val_accuracy: 0.5659
Epoch 27/250
202/202 [==============================] - 34s 171ms/step - loss: 1.0995 - accuracy: 0.5766 - val_loss: 1.1253 - val_accuracy: 0.5854
Epoch 28/250
202/202 [==============================] - 34s 168ms/step - loss: 1.0936 - accuracy: 0.5789 - val_loss: 1.1262 - val_accuracy: 0.5589
Epoch 29/250
202/202 [==============================] - 34s 169ms/step - loss: 1.0849 - accuracy: 0.5813 - val_loss: 1.1387 - val_accuracy: 0.5693
Epoch 30/250
202/202 [==============================] - 34s 170ms/step - loss: 1.0865 - accuracy: 0.5803 - val_loss: 1.1466 - val_accuracy: 0.5659
Epoch 31/250
202/202 [==============================] - 34s 169ms/step - loss: 1.0652 - accuracy: 0.5895 - val_loss: 1.0987 - val_accuracy: 0.5805
Epoch 32/250
202/202 [==============================] - 34s 168ms/step - loss: 1.0558 - accuracy: 0.5927 - val_loss: 1.1030 - val_accuracy: 0.5854
Epoch 33/250
202/202 [==============================] - 34s 168ms/step - loss: 1.0618 - accuracy: 0.5913 - val_loss: 1.1120 - val_accuracy: 0.5735
Epoch 34/250
202/202 [==============================] - 34s 170ms/step - loss: 1.0588 - accuracy: 0.5951 - val_loss: 1.0900 - val_accuracy: 0.5937
Epoch 35/250
202/202 [==============================] - 34s 169ms/step - loss: 0.9991 - accuracy: 0.6150 - val_loss: 1.0877 - val_accuracy: 0.5944
Epoch 36/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9832 - accuracy: 0.6194 - val_loss: 1.0823 - val_accuracy: 0.5937
Epoch 37/250
202/202 [==============================] - 34s 170ms/step - loss: 0.9742 - accuracy: 0.6276 - val_loss: 1.0661 - val_accuracy: 0.5875
Epoch 38/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9671 - accuracy: 0.6252 - val_loss: 1.0708 - val_accuracy: 0.5916
Epoch 39/250
202/202 [==============================] - 34s 169ms/step - loss: 0.9674 - accuracy: 0.6271 - val_loss: 1.0779 - val_accuracy: 0.5902
Epoch 40/250
202/202 [==============================] - 34s 170ms/step - loss: 0.9625 - accuracy: 0.6326 - val_loss: 1.0592 - val_accuracy: 0.6091
Epoch 41/250
202/202 [==============================] - 34s 170ms/step - loss: 0.9605 - accuracy: 0.6327 - val_loss: 1.0584 - val_accuracy: 0.5965
Epoch 42/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9527 - accuracy: 0.6316 - val_loss: 1.0627 - val_accuracy: 0.5958
Epoch 43/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9537 - accuracy: 0.6325 - val_loss: 1.0758 - val_accuracy: 0.5937
Epoch 44/250
202/202 [==============================] - 34s 169ms/step - loss: 0.9449 - accuracy: 0.6372 - val_loss: 1.0663 - val_accuracy: 0.6014
Epoch 45/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9399 - accuracy: 0.6422 - val_loss: 1.0728 - val_accuracy: 0.5979
Epoch 46/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9341 - accuracy: 0.6438 - val_loss: 1.0723 - val_accuracy: 0.6014
Epoch 47/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9292 - accuracy: 0.6406 - val_loss: 1.0849 - val_accuracy: 0.5986
Epoch 48/250
202/202 [==============================] - 36s 180ms/step - loss: 0.9227 - accuracy: 0.6453 - val_loss: 1.0800 - val_accuracy: 0.5944
Epoch 49/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9241 - accuracy: 0.6493 - val_loss: 1.0981 - val_accuracy: 0.5923
Epoch 50/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9136 - accuracy: 0.6503 - val_loss: 1.0756 - val_accuracy: 0.5958
Epoch 51/250
202/202 [==============================] - 34s 170ms/step - loss: 0.9048 - accuracy: 0.6540 - val_loss: 1.0591 - val_accuracy: 0.6000
Epoch 52/250
202/202 [==============================] - 34s 168ms/step - loss: 0.9056 - accuracy: 0.6496 - val_loss: 1.0770 - val_accuracy: 0.5909
Epoch 53/250
202/202 [==============================] - 34s 168ms/step - loss: 0.8901 - accuracy: 0.6579 - val_loss: 1.1022 - val_accuracy: 0.5930
Epoch 54/250
202/202 [==============================] - 34s 167ms/step - loss: 0.8934 - accuracy: 0.6547 - val_loss: 1.0798 - val_accuracy: 0.6000
Epoch 55/250
202/202 [==============================] - 34s 170ms/step - loss: 0.8872 - accuracy: 0.6626 - val_loss: 1.0876 - val_accuracy: 0.5993
Epoch 56/250
202/202 [==============================] - 34s 167ms/step - loss: 0.8870 - accuracy: 0.6604 - val_loss: 1.0697 - val_accuracy: 0.5979
Epoch 57/250
202/202 [==============================] - 34s 168ms/step - loss: 0.8852 - accuracy: 0.6627 - val_loss: 1.0648 - val_accuracy: 0.5993
Epoch 58/250
202/202 [==============================] - 34s 169ms/step - loss: 0.8754 - accuracy: 0.6636 - val_loss: 1.0827 - val_accuracy: 0.6035
Epoch 59/250
202/202 [==============================] - 34s 168ms/step - loss: 0.8654 - accuracy: 0.6674 - val_loss: 1.0856 - val_accuracy: 0.6000
Epoch 60/250
202/202 [==============================] - 34s 167ms/step - loss: 0.8720 - accuracy: 0.6678 - val_loss: 1.0895 - val_accuracy: 0.5986
Epoch 61/250
202/202 [==============================] - 34s 167ms/step - loss: 0.8617 - accuracy: 0.6732 - val_loss: 1.1048 - val_accuracy: 0.6049
Epoch 62/250
202/202 [==============================] - 34s 170ms/step - loss: 0.8552 - accuracy: 0.6739 - val_loss: 1.0932 - val_accuracy: 0.6014
Epoch 63/250
202/202 [==============================] - 34s 168ms/step - loss: 0.8531 - accuracy: 0.6706 - val_loss: 1.0755 - val_accuracy: 0.6035
Epoch 64/250
202/202 [==============================] - 34s 167ms/step - loss: 0.8532 - accuracy: 0.6757 - val_loss: 1.0691 - val_accuracy: 0.5986
Epoch 65/250
202/202 [==============================] - 34s 169ms/step - loss: 0.8537 - accuracy: 0.6739 - val_loss: 1.0902 - val_accuracy: 0.6077
Epoch 66/250
202/202 [==============================] - 34s 169ms/step - loss: 0.8150 - accuracy: 0.6907 - val_loss: 1.0887 - val_accuracy: 0.6035
Epoch 67/250
202/202 [==============================] - 34s 170ms/step - loss: 0.8080 - accuracy: 0.6975 - val_loss: 1.0667 - val_accuracy: 0.6084
Epoch 68/250
202/202 [==============================] - 34s 167ms/step - loss: 0.8013 - accuracy: 0.6955 - val_loss: 1.0846 - val_accuracy: 0.6028
Epoch 69/250
202/202 [==============================] - 34s 169ms/step - loss: 0.7901 - accuracy: 0.6979 - val_loss: 1.1064 - val_accuracy: 0.5986
Epoch 70/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7909 - accuracy: 0.6960 - val_loss: 1.1054 - val_accuracy: 0.6007
Epoch 71/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7879 - accuracy: 0.6988 - val_loss: 1.1090 - val_accuracy: 0.6014
Epoch 72/250
202/202 [==============================] - 35s 171ms/step - loss: 0.7946 - accuracy: 0.7000 - val_loss: 1.1111 - val_accuracy: 0.6049
Epoch 73/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7867 - accuracy: 0.6986 - val_loss: 1.1116 - val_accuracy: 0.5951
Epoch 74/250
202/202 [==============================] - 35s 171ms/step - loss: 0.7829 - accuracy: 0.7030 - val_loss: 1.0958 - val_accuracy: 0.6035
Epoch 75/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7784 - accuracy: 0.7019 - val_loss: 1.0953 - val_accuracy: 0.5993
Epoch 76/250
202/202 [==============================] - 34s 170ms/step - loss: 0.7739 - accuracy: 0.7032 - val_loss: 1.1242 - val_accuracy: 0.6000
Epoch 77/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7755 - accuracy: 0.7062 - val_loss: 1.1054 - val_accuracy: 0.6042
Epoch 78/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7744 - accuracy: 0.7050 - val_loss: 1.1337 - val_accuracy: 0.6021
Epoch 79/250
202/202 [==============================] - 34s 170ms/step - loss: 0.7633 - accuracy: 0.7106 - val_loss: 1.1081 - val_accuracy: 0.6021
Epoch 80/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7557 - accuracy: 0.7178 - val_loss: 1.1140 - val_accuracy: 0.6105
Epoch 81/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7523 - accuracy: 0.7170 - val_loss: 1.1059 - val_accuracy: 0.6077
Epoch 82/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7563 - accuracy: 0.7125 - val_loss: 1.1066 - val_accuracy: 0.6042
Epoch 83/250
202/202 [==============================] - 34s 169ms/step - loss: 0.7506 - accuracy: 0.7148 - val_loss: 1.1168 - val_accuracy: 0.6021
Epoch 84/250
202/202 [==============================] - 38s 188ms/step - loss: 0.7541 - accuracy: 0.7130 - val_loss: 1.1192 - val_accuracy: 0.6042
Epoch 85/250
202/202 [==============================] - 42s 210ms/step - loss: 0.7486 - accuracy: 0.7129 - val_loss: 1.1122 - val_accuracy: 0.6077
Epoch 86/250
202/202 [==============================] - 49s 242ms/step - loss: 0.7509 - accuracy: 0.7164 - val_loss: 1.1242 - val_accuracy: 0.6035
Epoch 87/250
202/202 [==============================] - 43s 214ms/step - loss: 0.7521 - accuracy: 0.7132 - val_loss: 1.1166 - val_accuracy: 0.6021
Epoch 88/250
202/202 [==============================] - 43s 212ms/step - loss: 0.7385 - accuracy: 0.7220 - val_loss: 1.1190 - val_accuracy: 0.6070
Epoch 89/250
202/202 [==============================] - 40s 200ms/step - loss: 0.7378 - accuracy: 0.7206 - val_loss: 1.1131 - val_accuracy: 0.6049
Epoch 90/250
202/202 [==============================] - 45s 223ms/step - loss: 0.7400 - accuracy: 0.7201 - val_loss: 1.1194 - val_accuracy: 0.6007
Epoch 91/250
202/202 [==============================] - 44s 218ms/step - loss: 0.7414 - accuracy: 0.7172 - val_loss: 1.1150 - val_accuracy: 0.6063
Epoch 92/250
202/202 [==============================] - 45s 224ms/step - loss: 0.7399 - accuracy: 0.7182 - val_loss: 1.1195 - val_accuracy: 0.60560.7396 - accuracy: 0.
Epoch 93/250
202/202 [==============================] - 43s 211ms/step - loss: 0.7371 - accuracy: 0.7244 - val_loss: 1.1204 - val_accuracy: 0.6063
Epoch 94/250
202/202 [==============================] - 39s 194ms/step - loss: 0.7356 - accuracy: 0.7244 - val_loss: 1.1176 - val_accuracy: 0.6049
Epoch 95/250
202/202 [==============================] - 43s 213ms/step - loss: 0.7341 - accuracy: 0.7197 - val_loss: 1.1153 - val_accuracy: 0.6077
Epoch 96/250
202/202 [==============================] - 46s 227ms/step - loss: 0.7380 - accuracy: 0.7215 - val_loss: 1.1142 - val_accuracy: 0.6063
Epoch 97/250
202/202 [==============================] - 41s 204ms/step - loss: 0.7348 - accuracy: 0.7249 - val_loss: 1.1169 - val_accuracy: 0.6063
Epoch 98/250
202/202 [==============================] - 39s 192ms/step - loss: 0.7353 - accuracy: 0.7226 - val_loss: 1.1171 - val_accuracy: 0.6084
Epoch 99/250
202/202 [==============================] - 39s 195ms/step - loss: 0.7306 - accuracy: 0.7237 - val_loss: 1.1196 - val_accuracy: 0.6077
Epoch 100/250
202/202 [==============================] - 39s 194ms/step - loss: 0.7305 - accuracy: 0.7223 - val_loss: 1.1187 - val_accuracy: 0.6077
Epoch 101/250
202/202 [==============================] - 38s 188ms/step - loss: 0.7337 - accuracy: 0.7253 - val_loss: 1.1180 - val_accuracy: 0.6098
Epoch 102/250
202/202 [==============================] - 41s 205ms/step - loss: 0.7301 - accuracy: 0.7239 - val_loss: 1.1183 - val_accuracy: 0.6084
Epoch 103/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7330 - accuracy: 0.7238 - val_loss: 1.1184 - val_accuracy: 0.6091
Epoch 104/250
202/202 [==============================] - 35s 175ms/step - loss: 0.7351 - accuracy: 0.7232 - val_loss: 1.1180 - val_accuracy: 0.6091
Epoch 105/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7299 - accuracy: 0.7275 - val_loss: 1.1184 - val_accuracy: 0.6091
Epoch 106/250
202/202 [==============================] - 37s 183ms/step - loss: 0.7344 - accuracy: 0.7226 - val_loss: 1.1186 - val_accuracy: 0.6091
Epoch 107/250
202/202 [==============================] - 37s 184ms/step - loss: 0.7378 - accuracy: 0.7207 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 108/250
202/202 [==============================] - 36s 179ms/step - loss: 0.7369 - accuracy: 0.7242 - val_loss: 1.1186 - val_accuracy: 0.6091
Epoch 109/250
202/202 [==============================] - 42s 210ms/step - loss: 0.7350 - accuracy: 0.7237 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 110/250
202/202 [==============================] - 38s 188ms/step - loss: 0.7294 - accuracy: 0.7250 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 111/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7314 - accuracy: 0.7235 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 112/250
202/202 [==============================] - 35s 174ms/step - loss: 0.7324 - accuracy: 0.7248 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 113/250
202/202 [==============================] - 37s 183ms/step - loss: 0.7285 - accuracy: 0.7252 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 114/250
202/202 [==============================] - 36s 177ms/step - loss: 0.7331 - accuracy: 0.7220 - val_loss: 1.1185 - val_accuracy: 0.6091
Epoch 115/250
202/202 [==============================] - 48s 237ms/step - loss: 0.7298 - accuracy: 0.7226 - val_loss: 1.1186 - val_accuracy: 0.6091
Epoch 116/250
202/202 [==============================] - 43s 215ms/step - loss: 0.7293 - accuracy: 0.7233 - val_loss: 1.1186 - val_accuracy: 0.6091
Epoch 117/250
202/202 [==============================] - 53s 261ms/step - loss: 0.7333 - accuracy: 0.7269 - val_loss: 1.1186 - val_accuracy: 0.6091
Epoch 118/250
202/202 [==============================] - 65s 323ms/step - loss: 0.7381 - accuracy: 0.7228 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 119/250
202/202 [==============================] - 44s 217ms/step - loss: 0.7316 - accuracy: 0.7248 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 120/250
202/202 [==============================] - 52s 259ms/step - loss: 0.7331 - accuracy: 0.7208 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 121/250
202/202 [==============================] - 44s 219ms/step - loss: 0.7297 - accuracy: 0.7269 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 122/250
202/202 [==============================] - 37s 182ms/step - loss: 0.7357 - accuracy: 0.7202 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 123/250
202/202 [==============================] - 36s 179ms/step - loss: 0.7351 - accuracy: 0.7230 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 124/250
202/202 [==============================] - 37s 181ms/step - loss: 0.7292 - accuracy: 0.7248 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 125/250
202/202 [==============================] - 37s 181ms/step - loss: 0.7326 - accuracy: 0.7236 - val_loss: 1.1186 - val_accuracy: 0.6091
Epoch 126/250
202/202 [==============================] - 37s 181ms/step - loss: 0.7279 - accuracy: 0.7259 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 127/250
202/202 [==============================] - 36s 180ms/step - loss: 0.7349 - accuracy: 0.7230 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 128/250
202/202 [==============================] - 36s 180ms/step - loss: 0.7307 - accuracy: 0.7261 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 129/250
202/202 [==============================] - 36s 181ms/step - loss: 0.7323 - accuracy: 0.7230 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 130/250
202/202 [==============================] - 37s 182ms/step - loss: 0.7423 - accuracy: 0.7196 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 131/250
202/202 [==============================] - 36s 179ms/step - loss: 0.7319 - accuracy: 0.7241 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 132/250
202/202 [==============================] - 37s 181ms/step - loss: 0.7296 - accuracy: 0.7258 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 133/250
202/202 [==============================] - 36s 179ms/step - loss: 0.7326 - accuracy: 0.7231 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 134/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7352 - accuracy: 0.7241 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 135/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7326 - accuracy: 0.7199 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 136/250
202/202 [==============================] - 36s 180ms/step - loss: 0.7311 - accuracy: 0.7243 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 137/250
202/202 [==============================] - 37s 184ms/step - loss: 0.7277 - accuracy: 0.7236 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 138/250
202/202 [==============================] - 48s 240ms/step - loss: 0.7304 - accuracy: 0.7265 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 139/250
202/202 [==============================] - 47s 234ms/step - loss: 0.7328 - accuracy: 0.7250 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 140/250
202/202 [==============================] - 47s 232ms/step - loss: 0.7371 - accuracy: 0.7236 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 141/250
202/202 [==============================] - 47s 232ms/step - loss: 0.7332 - accuracy: 0.7197 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 142/250
202/202 [==============================] - 49s 243ms/step - loss: 0.7284 - accuracy: 0.7232 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 143/250
202/202 [==============================] - 48s 237ms/step - loss: 0.7337 - accuracy: 0.7245 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 144/250
202/202 [==============================] - 44s 219ms/step - loss: 0.7286 - accuracy: 0.7254 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 145/250
202/202 [==============================] - 38s 186ms/step - loss: 0.7277 - accuracy: 0.7250 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 146/250
202/202 [==============================] - 44s 218ms/step - loss: 0.7289 - accuracy: 0.7257 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 147/250
202/202 [==============================] - 41s 202ms/step - loss: 0.7313 - accuracy: 0.7206 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 148/250
202/202 [==============================] - 42s 206ms/step - loss: 0.7353 - accuracy: 0.7216 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 149/250
202/202 [==============================] - 41s 201ms/step - loss: 0.7353 - accuracy: 0.7223 - val_loss: 1.1187 - val_accuracy: 0.6091
Epoch 150/250
202/202 [==============================] - 41s 204ms/step - loss: 0.7298 - accuracy: 0.7264 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 151/250
202/202 [==============================] - 39s 192ms/step - loss: 0.7314 - accuracy: 0.7216 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 152/250
202/202 [==============================] - 41s 204ms/step - loss: 0.7324 - accuracy: 0.7285 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 153/250
202/202 [==============================] - 57s 283ms/step - loss: 0.7310 - accuracy: 0.7240 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 154/250
202/202 [==============================] - 69s 339ms/step - loss: 0.7303 - accuracy: 0.7230 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 155/250
202/202 [==============================] - 62s 305ms/step - loss: 0.7299 - accuracy: 0.7219 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 156/250
202/202 [==============================] - 68s 336ms/step - loss: 0.7305 - accuracy: 0.7253 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 157/250
202/202 [==============================] - 96s 474ms/step - loss: 0.7271 - accuracy: 0.7261 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 158/250
202/202 [==============================] - 90s 448ms/step - loss: 0.7324 - accuracy: 0.7262 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 159/250
202/202 [==============================] - 64s 319ms/step - loss: 0.7352 - accuracy: 0.7240 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 160/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7347 - accuracy: 0.7250 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 161/250
202/202 [==============================] - 36s 177ms/step - loss: 0.7222 - accuracy: 0.7285 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 162/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7352 - accuracy: 0.7245 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 163/250
202/202 [==============================] - 36s 176ms/step - loss: 0.7388 - accuracy: 0.7199 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 164/250
202/202 [==============================] - 36s 177ms/step - loss: 0.7303 - accuracy: 0.7254 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 165/250
202/202 [==============================] - 36s 179ms/step - loss: 0.7301 - accuracy: 0.7206 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 166/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7283 - accuracy: 0.7233 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 167/250
202/202 [==============================] - 36s 177ms/step - loss: 0.7319 - accuracy: 0.7252 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 168/250
202/202 [==============================] - 36s 179ms/step - loss: 0.7387 - accuracy: 0.7185 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 169/250
202/202 [==============================] - 36s 176ms/step - loss: 0.7366 - accuracy: 0.7225 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 170/250
202/202 [==============================] - 35s 175ms/step - loss: 0.7308 - accuracy: 0.7253 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 171/250
202/202 [==============================] - 35s 174ms/step - loss: 0.7261 - accuracy: 0.7281 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 172/250
202/202 [==============================] - 36s 176ms/step - loss: 0.7311 - accuracy: 0.7247 - val_loss: 1.1190 - val_accuracy: 0.6091
Epoch 173/250
202/202 [==============================] - 35s 174ms/step - loss: 0.7326 - accuracy: 0.7226 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 174/250
202/202 [==============================] - 35s 174ms/step - loss: 0.7315 - accuracy: 0.7248 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 175/250
202/202 [==============================] - 35s 176ms/step - loss: 0.7303 - accuracy: 0.7225 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 176/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7304 - accuracy: 0.7265 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 177/250
202/202 [==============================] - 36s 177ms/step - loss: 0.7373 - accuracy: 0.7267 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 178/250
202/202 [==============================] - 35s 174ms/step - loss: 0.7262 - accuracy: 0.7279 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 179/250
202/202 [==============================] - 35s 175ms/step - loss: 0.7293 - accuracy: 0.7274 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 180/250
202/202 [==============================] - 35s 174ms/step - loss: 0.7305 - accuracy: 0.7244 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 181/250
202/202 [==============================] - 35s 173ms/step - loss: 0.7355 - accuracy: 0.7221 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 182/250
202/202 [==============================] - 35s 176ms/step - loss: 0.7353 - accuracy: 0.7219 - val_loss: 1.1187 - val_accuracy: 0.6084
Epoch 183/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7281 - accuracy: 0.7249 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 184/250
202/202 [==============================] - 35s 175ms/step - loss: 0.7318 - accuracy: 0.7244 - val_loss: 1.1187 - val_accuracy: 0.6084
Epoch 185/250
202/202 [==============================] - 35s 172ms/step - loss: 0.7268 - accuracy: 0.7231 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 186/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7326 - accuracy: 0.7271 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 187/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7344 - accuracy: 0.7261 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 188/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7358 - accuracy: 0.7203 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 189/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7280 - accuracy: 0.7214 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 190/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7282 - accuracy: 0.7262 - val_loss: 1.1188 - val_accuracy: 0.6091
Epoch 191/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7282 - accuracy: 0.7274 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 192/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7343 - accuracy: 0.7260 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 193/250
202/202 [==============================] - 34s 169ms/step - loss: 0.7295 - accuracy: 0.7233 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 194/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7321 - accuracy: 0.7261 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 195/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7285 - accuracy: 0.7245 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 196/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7357 - accuracy: 0.7222 - val_loss: 1.1189 - val_accuracy: 0.6091
Epoch 197/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7342 - accuracy: 0.7239 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 198/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7317 - accuracy: 0.7261 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 199/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7292 - accuracy: 0.7266 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 200/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7264 - accuracy: 0.7247 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 201/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7323 - accuracy: 0.7252 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 202/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7289 - accuracy: 0.7233 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 203/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7316 - accuracy: 0.7249 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 204/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7375 - accuracy: 0.7208 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 205/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7326 - accuracy: 0.7221 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 206/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7315 - accuracy: 0.7261 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 207/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7283 - accuracy: 0.7299 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 208/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7412 - accuracy: 0.7243 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 209/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7399 - accuracy: 0.7218 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 210/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7303 - accuracy: 0.7237 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 211/250
202/202 [==============================] - 34s 169ms/step - loss: 0.7333 - accuracy: 0.7239 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 212/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7313 - accuracy: 0.7233 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 213/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7304 - accuracy: 0.7233 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 214/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7321 - accuracy: 0.7224 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 215/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7302 - accuracy: 0.7237 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 216/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7317 - accuracy: 0.7268 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 217/250
202/202 [==============================] - 34s 169ms/step - loss: 0.7332 - accuracy: 0.7246 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 218/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7361 - accuracy: 0.7228 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 219/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7330 - accuracy: 0.7240 - val_loss: 1.1190 - val_accuracy: 0.6077
Epoch 220/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7318 - accuracy: 0.7254 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 221/250
202/202 [==============================] - 34s 166ms/step - loss: 0.7287 - accuracy: 0.7277 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 222/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7336 - accuracy: 0.7239 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 223/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7312 - accuracy: 0.7277 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 224/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7325 - accuracy: 0.7258 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 225/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7313 - accuracy: 0.7259 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 226/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7333 - accuracy: 0.7242 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 227/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7310 - accuracy: 0.7241 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 228/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7354 - accuracy: 0.7238 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 229/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7259 - accuracy: 0.7280 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 230/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7310 - accuracy: 0.7242 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 231/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7273 - accuracy: 0.7256 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 232/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7369 - accuracy: 0.7226 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 233/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7337 - accuracy: 0.7247 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 234/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7337 - accuracy: 0.7216 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 235/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7323 - accuracy: 0.7261 - val_loss: 1.1187 - val_accuracy: 0.6077
Epoch 236/250
202/202 [==============================] - 36s 178ms/step - loss: 0.7276 - accuracy: 0.7254 - val_loss: 1.1187 - val_accuracy: 0.6077
Epoch 237/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7228 - accuracy: 0.7264 - val_loss: 1.1187 - val_accuracy: 0.6077
Epoch 238/250
202/202 [==============================] - 33s 166ms/step - loss: 0.7320 - accuracy: 0.7242 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 239/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7266 - accuracy: 0.7294 - val_loss: 1.1189 - val_accuracy: 0.6084
Epoch 240/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7348 - accuracy: 0.7230 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 241/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7307 - accuracy: 0.7220 - val_loss: 1.1188 - val_accuracy: 0.6084
Epoch 242/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7278 - accuracy: 0.7259 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 243/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7313 - accuracy: 0.7247 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 244/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7327 - accuracy: 0.7240 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 245/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7312 - accuracy: 0.7270 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 246/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7289 - accuracy: 0.7239 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 247/250
202/202 [==============================] - 34s 168ms/step - loss: 0.7340 - accuracy: 0.7242 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 248/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7293 - accuracy: 0.7256 - val_loss: 1.1189 - val_accuracy: 0.6077
Epoch 249/250
202/202 [==============================] - 33s 165ms/step - loss: 0.7264 - accuracy: 0.7277 - val_loss: 1.1188 - val_accuracy: 0.6077
Epoch 250/250
202/202 [==============================] - 34s 167ms/step - loss: 0.7299 - accuracy: 0.7268 - val_loss: 1.1188 - val_accuracy: 0.6077

Step 6.2.2: Learning Curve

plt.style.use('fivethirtyeight')
epochs = [i for i in range(250)]
fig , ax = plt.subplots(1,2)
train_acc = history_1_e.history['accuracy']
train_loss = history_1_e.history['loss']
test_acc = history_1_e.history['val_accuracy']
test_loss = history_1_e.history['val_loss']

fig.set_size_inches(20,6)
ax[0].plot(epochs , train_loss , label = 'Training Loss')
ax[0].plot(epochs , test_loss , label = 'Testing Loss')
ax[0].set_title('Training & Testing Loss')
ax[0].legend()
ax[0].set_xlabel("Epochs")

ax[1].plot(epochs , train_acc , label = 'Training Accuracy')
ax[1].plot(epochs , test_acc , label = 'Testing Accuracy')
ax[1].set_title('Training & Testing Accuracy')
ax[1].legend()
ax[1].set_xlabel("Epochs")
plt.show()

Step 6.2.3: Save Model

# Save Model

model_1_e.save('Trained Models/model_1_e.h5')

Step 7: Train LSTM Classifier

Step 7.1: For Gender Classification

x_train, x_test, y_train, y_test = X_train, X_test, y1_train, y1_test

size = 13
x_train = np.array(x_train)
y_train = np.array(y_train)
x_test = np.array(x_test) 
y_test = np.array(y_test)
x_train = x_train.reshape(x_train.shape[0], size, 1)
x_test = x_test.reshape(x_test.shape[0], size, 1)
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)

# Create Model

from tensorflow.keras.layers import LSTM, GRU, Dense, Embedding, Dropout, GlobalAveragePooling1D, Flatten, SpatialDropout1D, Bidirectional

lstm_1_g = Sequential()
lstm_1_g.add(Dense(256,  activation='relu', input_shape=(x_train.shape[1],1)))
lstm_1_g.add(Dense(128, activation='relu'))
lstm_1_g.add(Dense(64, activation='relu'))
lstm_1_g.add(Dense(32, activation='relu'))
lstm_1_g.add(Dense(16, activation='relu'))
lstm_1_g.add(Dense(8, activation='relu'))
lstm_1_g.add(Flatten())
lstm_1_g.add(Dense(2, activation='softmax'))
lstm_1_g.compile(optimizer = 'adam' , loss = 'categorical_crossentropy' , metrics = ['accuracy'])
lstm_1_g.summary()

Model: "sequential_2"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_4 (Dense)              (None, 13, 256)           512       
_________________________________________________________________
dense_5 (Dense)              (None, 13, 128)           32896     
_________________________________________________________________
dense_6 (Dense)              (None, 13, 64)            8256      
_________________________________________________________________
dense_7 (Dense)              (None, 13, 32)            2080      
_________________________________________________________________
dense_8 (Dense)              (None, 13, 16)            528       
_________________________________________________________________
dense_9 (Dense)              (None, 13, 8)             136       
_________________________________________________________________
flatten_2 (Flatten)          (None, 104)               0         
_________________________________________________________________
dense_10 (Dense)             (None, 2)                 210       
=================================================================
Total params: 44,618
Trainable params: 44,618
Non-trainable params: 0
_________________________________________________________________

Step 7.1.1: Training

history_2_g = lstm_1_g.fit(x_train, y_train, epochs=250, batch_size=64, validation_split=0.2)

Epoch 1/250
162/162 [==============================] - 4s 27ms/step - loss: 0.6324 - accuracy: 0.6241 - val_loss: 0.5345 - val_accuracy: 0.7147
Epoch 2/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4967 - accuracy: 0.7427 - val_loss: 0.5070 - val_accuracy: 0.7387
Epoch 3/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4776 - accuracy: 0.7548 - val_loss: 0.4952 - val_accuracy: 0.7433
Epoch 4/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4692 - accuracy: 0.7640 - val_loss: 0.5025 - val_accuracy: 0.7507
Epoch 5/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4661 - accuracy: 0.7692 - val_loss: 0.4851 - val_accuracy: 0.7584
Epoch 6/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4570 - accuracy: 0.7749 - val_loss: 0.4729 - val_accuracy: 0.7573
Epoch 7/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4532 - accuracy: 0.7798 - val_loss: 0.4663 - val_accuracy: 0.7642
Epoch 8/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4501 - accuracy: 0.7831 - val_loss: 0.4701 - val_accuracy: 0.7716
Epoch 9/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4488 - accuracy: 0.7814 - val_loss: 0.4693 - val_accuracy: 0.7727
Epoch 10/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4437 - accuracy: 0.7859 - val_loss: 0.4575 - val_accuracy: 0.7774
Epoch 11/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4408 - accuracy: 0.7870 - val_loss: 0.4701 - val_accuracy: 0.7600
Epoch 12/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4368 - accuracy: 0.7912 - val_loss: 0.4550 - val_accuracy: 0.7758
Epoch 13/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4375 - accuracy: 0.7913 - val_loss: 0.4509 - val_accuracy: 0.7762
Epoch 14/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4357 - accuracy: 0.7907 - val_loss: 0.4444 - val_accuracy: 0.7805
Epoch 15/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4331 - accuracy: 0.7906 - val_loss: 0.4447 - val_accuracy: 0.7801
Epoch 16/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4313 - accuracy: 0.7938 - val_loss: 0.4464 - val_accuracy: 0.7805
Epoch 17/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4306 - accuracy: 0.7929 - val_loss: 0.4563 - val_accuracy: 0.7743
Epoch 18/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4257 - accuracy: 0.7953 - val_loss: 0.4409 - val_accuracy: 0.7758
Epoch 19/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4275 - accuracy: 0.7929 - val_loss: 0.4348 - val_accuracy: 0.7840
Epoch 20/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4238 - accuracy: 0.7967 - val_loss: 0.4501 - val_accuracy: 0.7844
Epoch 21/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4217 - accuracy: 0.7968 - val_loss: 0.4417 - val_accuracy: 0.7801
Epoch 22/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4213 - accuracy: 0.7960 - val_loss: 0.4334 - val_accuracy: 0.7782
Epoch 23/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4201 - accuracy: 0.7968 - val_loss: 0.4401 - val_accuracy: 0.7836
Epoch 24/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4214 - accuracy: 0.7961 - val_loss: 0.4445 - val_accuracy: 0.7782
Epoch 25/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4196 - accuracy: 0.7976 - val_loss: 0.4317 - val_accuracy: 0.7789
Epoch 26/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4180 - accuracy: 0.7982 - val_loss: 0.4411 - val_accuracy: 0.7816
Epoch 27/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4180 - accuracy: 0.8003 - val_loss: 0.4348 - val_accuracy: 0.7863
Epoch 28/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4157 - accuracy: 0.7989 - val_loss: 0.4284 - val_accuracy: 0.7871
Epoch 29/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4129 - accuracy: 0.8019 - val_loss: 0.4353 - val_accuracy: 0.7875
Epoch 30/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4162 - accuracy: 0.8003 - val_loss: 0.4286 - val_accuracy: 0.7844
Epoch 31/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4157 - accuracy: 0.7967 - val_loss: 0.4300 - val_accuracy: 0.7844
Epoch 32/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4126 - accuracy: 0.8027 - val_loss: 0.4431 - val_accuracy: 0.7782
Epoch 33/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4119 - accuracy: 0.8002 - val_loss: 0.4278 - val_accuracy: 0.7906
Epoch 34/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4093 - accuracy: 0.8036 - val_loss: 0.4288 - val_accuracy: 0.7886
Epoch 35/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4086 - accuracy: 0.8038 - val_loss: 0.4232 - val_accuracy: 0.7898
Epoch 36/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4121 - accuracy: 0.8008 - val_loss: 0.4419 - val_accuracy: 0.7789
Epoch 37/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4103 - accuracy: 0.8008 - val_loss: 0.4306 - val_accuracy: 0.7894
Epoch 38/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4107 - accuracy: 0.8044 - val_loss: 0.4235 - val_accuracy: 0.7917
Epoch 39/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4084 - accuracy: 0.8033 - val_loss: 0.4294 - val_accuracy: 0.7851
Epoch 40/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4075 - accuracy: 0.8027 - val_loss: 0.4302 - val_accuracy: 0.7824
Epoch 41/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4119 - accuracy: 0.7998 - val_loss: 0.4222 - val_accuracy: 0.7909
Epoch 42/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4075 - accuracy: 0.8034 - val_loss: 0.4258 - val_accuracy: 0.7902
Epoch 43/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4112 - accuracy: 0.8040 - val_loss: 0.4244 - val_accuracy: 0.7886
Epoch 44/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4076 - accuracy: 0.8039 - val_loss: 0.4334 - val_accuracy: 0.7828
Epoch 45/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4057 - accuracy: 0.8062 - val_loss: 0.4207 - val_accuracy: 0.7890
Epoch 46/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4018 - accuracy: 0.8097 - val_loss: 0.4269 - val_accuracy: 0.7890
Epoch 47/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4058 - accuracy: 0.8027 - val_loss: 0.4257 - val_accuracy: 0.7933
Epoch 48/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4000 - accuracy: 0.8106 - val_loss: 0.4251 - val_accuracy: 0.7855
Epoch 49/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4071 - accuracy: 0.8025 - val_loss: 0.4520 - val_accuracy: 0.7704
Epoch 50/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4040 - accuracy: 0.8067 - val_loss: 0.4238 - val_accuracy: 0.7940
Epoch 51/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4001 - accuracy: 0.8074 - val_loss: 0.4195 - val_accuracy: 0.7886
Epoch 52/250
162/162 [==============================] - 4s 27ms/step - loss: 0.4037 - accuracy: 0.8084 - val_loss: 0.4175 - val_accuracy: 0.7902
Epoch 53/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4039 - accuracy: 0.8070 - val_loss: 0.4241 - val_accuracy: 0.7836
Epoch 54/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3991 - accuracy: 0.8112 - val_loss: 0.4267 - val_accuracy: 0.7998
Epoch 55/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4007 - accuracy: 0.8081 - val_loss: 0.4151 - val_accuracy: 0.7964
Epoch 56/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3973 - accuracy: 0.8106 - val_loss: 0.4164 - val_accuracy: 0.7960
Epoch 57/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3985 - accuracy: 0.8105 - val_loss: 0.4143 - val_accuracy: 0.7987
Epoch 58/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4025 - accuracy: 0.8071 - val_loss: 0.4205 - val_accuracy: 0.7952
Epoch 59/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3990 - accuracy: 0.8083 - val_loss: 0.4245 - val_accuracy: 0.7863
Epoch 60/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3965 - accuracy: 0.8101 - val_loss: 0.4198 - val_accuracy: 0.7983
Epoch 61/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3955 - accuracy: 0.8139 - val_loss: 0.4154 - val_accuracy: 0.7933
Epoch 62/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3963 - accuracy: 0.8117 - val_loss: 0.4095 - val_accuracy: 0.7975
Epoch 63/250
162/162 [==============================] - 4s 26ms/step - loss: 0.4003 - accuracy: 0.8075 - val_loss: 0.4083 - val_accuracy: 0.8026
Epoch 64/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3947 - accuracy: 0.8136 - val_loss: 0.4152 - val_accuracy: 0.7983
Epoch 65/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3999 - accuracy: 0.8086 - val_loss: 0.4078 - val_accuracy: 0.8002
Epoch 66/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3960 - accuracy: 0.8131 - val_loss: 0.4125 - val_accuracy: 0.7948
Epoch 67/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3950 - accuracy: 0.8120 - val_loss: 0.4120 - val_accuracy: 0.7925
Epoch 68/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3935 - accuracy: 0.8129 - val_loss: 0.4799 - val_accuracy: 0.7511
Epoch 69/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3991 - accuracy: 0.8086 - val_loss: 0.4083 - val_accuracy: 0.7998
Epoch 70/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3904 - accuracy: 0.8158 - val_loss: 0.4143 - val_accuracy: 0.8018
Epoch 71/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3960 - accuracy: 0.8104 - val_loss: 0.4040 - val_accuracy: 0.8033
Epoch 72/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3934 - accuracy: 0.8123 - val_loss: 0.4054 - val_accuracy: 0.7975
Epoch 73/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3918 - accuracy: 0.8141 - val_loss: 0.4080 - val_accuracy: 0.7991
Epoch 74/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3963 - accuracy: 0.8087 - val_loss: 0.4072 - val_accuracy: 0.8041
Epoch 75/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3940 - accuracy: 0.8123 - val_loss: 0.4069 - val_accuracy: 0.8041
Epoch 76/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3927 - accuracy: 0.8134 - val_loss: 0.4086 - val_accuracy: 0.8037
Epoch 77/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3947 - accuracy: 0.8107 - val_loss: 0.4236 - val_accuracy: 0.7971
Epoch 78/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3949 - accuracy: 0.8105 - val_loss: 0.4249 - val_accuracy: 0.7859
Epoch 79/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3930 - accuracy: 0.8156 - val_loss: 0.4347 - val_accuracy: 0.7917
Epoch 80/250
162/162 [==============================] - 4s 28ms/step - loss: 0.3992 - accuracy: 0.8119 - val_loss: 0.4057 - val_accuracy: 0.8010
Epoch 81/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3914 - accuracy: 0.8164 - val_loss: 0.4070 - val_accuracy: 0.8037
Epoch 82/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3889 - accuracy: 0.8152 - val_loss: 0.4214 - val_accuracy: 0.7925
Epoch 83/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3917 - accuracy: 0.8168 - val_loss: 0.4013 - val_accuracy: 0.8026
Epoch 84/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3951 - accuracy: 0.8135 - val_loss: 0.4014 - val_accuracy: 0.8018
Epoch 85/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3919 - accuracy: 0.8168 - val_loss: 0.4062 - val_accuracy: 0.8002
Epoch 86/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3924 - accuracy: 0.8134 - val_loss: 0.4167 - val_accuracy: 0.7929
Epoch 87/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3902 - accuracy: 0.8158 - val_loss: 0.4051 - val_accuracy: 0.8006
Epoch 88/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3907 - accuracy: 0.8176 - val_loss: 0.4224 - val_accuracy: 0.7933
Epoch 89/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3941 - accuracy: 0.8148 - val_loss: 0.4324 - val_accuracy: 0.7890
Epoch 90/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3984 - accuracy: 0.8109 - val_loss: 0.4727 - val_accuracy: 0.7782
Epoch 91/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3965 - accuracy: 0.8128 - val_loss: 0.4243 - val_accuracy: 0.7898
Epoch 92/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3964 - accuracy: 0.8084 - val_loss: 0.4148 - val_accuracy: 0.7956
Epoch 93/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3888 - accuracy: 0.8186 - val_loss: 0.4162 - val_accuracy: 0.7909
Epoch 94/250
162/162 [==============================] - 5s 30ms/step - loss: 0.3880 - accuracy: 0.8195 - val_loss: 0.4005 - val_accuracy: 0.8060
Epoch 95/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3887 - accuracy: 0.8174 - val_loss: 0.4142 - val_accuracy: 0.7940
Epoch 96/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3855 - accuracy: 0.8196 - val_loss: 0.4118 - val_accuracy: 0.7967
Epoch 97/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3966 - accuracy: 0.8099 - val_loss: 0.4057 - val_accuracy: 0.8049
Epoch 98/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3861 - accuracy: 0.8191 - val_loss: 0.4177 - val_accuracy: 0.8002
Epoch 99/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3962 - accuracy: 0.8096 - val_loss: 0.4136 - val_accuracy: 0.7956
Epoch 100/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3886 - accuracy: 0.8161 - val_loss: 0.4195 - val_accuracy: 0.7906
Epoch 101/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3925 - accuracy: 0.8165 - val_loss: 0.4016 - val_accuracy: 0.8026
Epoch 102/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3906 - accuracy: 0.8172 - val_loss: 0.4162 - val_accuracy: 0.8006
Epoch 103/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3883 - accuracy: 0.8192 - val_loss: 0.4019 - val_accuracy: 0.8057
Epoch 104/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3889 - accuracy: 0.8159 - val_loss: 0.4138 - val_accuracy: 0.7967
Epoch 105/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3883 - accuracy: 0.8139 - val_loss: 0.4088 - val_accuracy: 0.8041
Epoch 106/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3880 - accuracy: 0.8186 - val_loss: 0.4095 - val_accuracy: 0.8076
Epoch 107/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3912 - accuracy: 0.8126 - val_loss: 0.4268 - val_accuracy: 0.7882
Epoch 108/250
162/162 [==============================] - 4s 28ms/step - loss: 0.3892 - accuracy: 0.8199 - val_loss: 0.4015 - val_accuracy: 0.8014
Epoch 109/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3863 - accuracy: 0.8218 - val_loss: 0.4062 - val_accuracy: 0.8033
Epoch 110/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3866 - accuracy: 0.8176 - val_loss: 0.4243 - val_accuracy: 0.7902
Epoch 111/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3881 - accuracy: 0.8174 - val_loss: 0.4131 - val_accuracy: 0.8045
Epoch 112/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3860 - accuracy: 0.8225 - val_loss: 0.4104 - val_accuracy: 0.8022
Epoch 113/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3901 - accuracy: 0.8160 - val_loss: 0.4088 - val_accuracy: 0.7987
Epoch 114/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3867 - accuracy: 0.8190 - val_loss: 0.4141 - val_accuracy: 0.8006
Epoch 115/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3873 - accuracy: 0.8200 - val_loss: 0.4016 - val_accuracy: 0.8026
Epoch 116/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3901 - accuracy: 0.8186 - val_loss: 0.4182 - val_accuracy: 0.7944
Epoch 117/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3880 - accuracy: 0.8206 - val_loss: 0.4037 - val_accuracy: 0.8037
Epoch 118/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3875 - accuracy: 0.8194 - val_loss: 0.3996 - val_accuracy: 0.8060
Epoch 119/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3876 - accuracy: 0.8157 - val_loss: 0.3996 - val_accuracy: 0.8068
Epoch 120/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3882 - accuracy: 0.8156 - val_loss: 0.4022 - val_accuracy: 0.8014
Epoch 121/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3863 - accuracy: 0.8185 - val_loss: 0.4019 - val_accuracy: 0.8049
Epoch 122/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3874 - accuracy: 0.8190 - val_loss: 0.4050 - val_accuracy: 0.8057
Epoch 123/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3883 - accuracy: 0.8181 - val_loss: 0.4029 - val_accuracy: 0.8049
Epoch 124/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3895 - accuracy: 0.8164 - val_loss: 0.4010 - val_accuracy: 0.8018
Epoch 125/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3896 - accuracy: 0.8164 - val_loss: 0.4034 - val_accuracy: 0.8037
Epoch 126/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3837 - accuracy: 0.8206 - val_loss: 0.4147 - val_accuracy: 0.8049
Epoch 127/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3870 - accuracy: 0.8199 - val_loss: 0.3977 - val_accuracy: 0.8068
Epoch 128/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3835 - accuracy: 0.8220 - val_loss: 0.3957 - val_accuracy: 0.8057
Epoch 129/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3886 - accuracy: 0.8189 - val_loss: 0.4060 - val_accuracy: 0.8041
Epoch 130/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3863 - accuracy: 0.8185 - val_loss: 0.4068 - val_accuracy: 0.8006
Epoch 131/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3870 - accuracy: 0.8188 - val_loss: 0.4063 - val_accuracy: 0.8014
Epoch 132/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3856 - accuracy: 0.8228 - val_loss: 0.4647 - val_accuracy: 0.7716
Epoch 133/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3862 - accuracy: 0.8192 - val_loss: 0.3964 - val_accuracy: 0.8057
Epoch 134/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3819 - accuracy: 0.8231 - val_loss: 0.3994 - val_accuracy: 0.8107
Epoch 135/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3874 - accuracy: 0.8150 - val_loss: 0.4150 - val_accuracy: 0.8002
Epoch 136/250
162/162 [==============================] - 4s 28ms/step - loss: 0.3839 - accuracy: 0.8221 - val_loss: 0.3995 - val_accuracy: 0.8060
Epoch 137/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3852 - accuracy: 0.8175 - val_loss: 0.4168 - val_accuracy: 0.7964
Epoch 138/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3847 - accuracy: 0.8212 - val_loss: 0.4125 - val_accuracy: 0.7929
Epoch 139/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3862 - accuracy: 0.8182 - val_loss: 0.3982 - val_accuracy: 0.8111
Epoch 140/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3852 - accuracy: 0.8228 - val_loss: 0.4015 - val_accuracy: 0.8068
Epoch 141/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3866 - accuracy: 0.8211 - val_loss: 0.3996 - val_accuracy: 0.8057
Epoch 142/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3894 - accuracy: 0.8168 - val_loss: 0.3996 - val_accuracy: 0.8087
Epoch 143/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3835 - accuracy: 0.8222 - val_loss: 0.3975 - val_accuracy: 0.8076
Epoch 144/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3854 - accuracy: 0.8207 - val_loss: 0.4015 - val_accuracy: 0.8072
Epoch 145/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3821 - accuracy: 0.8245 - val_loss: 0.4040 - val_accuracy: 0.8080
Epoch 146/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3816 - accuracy: 0.8237 - val_loss: 0.4016 - val_accuracy: 0.8130
Epoch 147/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3832 - accuracy: 0.8216 - val_loss: 0.4053 - val_accuracy: 0.8087
Epoch 148/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3860 - accuracy: 0.8202 - val_loss: 0.4008 - val_accuracy: 0.8053
Epoch 149/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3856 - accuracy: 0.8219 - val_loss: 0.4006 - val_accuracy: 0.8091
Epoch 150/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3885 - accuracy: 0.8199 - val_loss: 0.4390 - val_accuracy: 0.7840
Epoch 151/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3875 - accuracy: 0.8175 - val_loss: 0.4416 - val_accuracy: 0.7801
Epoch 152/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3857 - accuracy: 0.8187 - val_loss: 0.4127 - val_accuracy: 0.7956
Epoch 153/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3905 - accuracy: 0.8166 - val_loss: 0.4128 - val_accuracy: 0.7979
Epoch 154/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3845 - accuracy: 0.8206 - val_loss: 0.4064 - val_accuracy: 0.8010
Epoch 155/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3847 - accuracy: 0.8209 - val_loss: 0.4100 - val_accuracy: 0.8022
Epoch 156/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3847 - accuracy: 0.8213 - val_loss: 0.4073 - val_accuracy: 0.8115
Epoch 157/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3868 - accuracy: 0.8174 - val_loss: 0.4090 - val_accuracy: 0.8057
Epoch 158/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3863 - accuracy: 0.8183 - val_loss: 0.3993 - val_accuracy: 0.8018
Epoch 159/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3855 - accuracy: 0.8199 - val_loss: 0.4265 - val_accuracy: 0.7952
Epoch 160/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3877 - accuracy: 0.8205 - val_loss: 0.3984 - val_accuracy: 0.8049
Epoch 161/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3890 - accuracy: 0.8228 - val_loss: 0.4074 - val_accuracy: 0.8026
Epoch 162/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3840 - accuracy: 0.8202 - val_loss: 0.3998 - val_accuracy: 0.8033
Epoch 163/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3856 - accuracy: 0.8183 - val_loss: 0.4067 - val_accuracy: 0.8022
Epoch 164/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3926 - accuracy: 0.8143 - val_loss: 0.3997 - val_accuracy: 0.8057
Epoch 165/250
162/162 [==============================] - 4s 28ms/step - loss: 0.3840 - accuracy: 0.8205 - val_loss: 0.4043 - val_accuracy: 0.8068
Epoch 166/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3853 - accuracy: 0.8208 - val_loss: 0.4020 - val_accuracy: 0.8014
Epoch 167/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3852 - accuracy: 0.8149 - val_loss: 0.3946 - val_accuracy: 0.8084
Epoch 168/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3836 - accuracy: 0.8196 - val_loss: 0.3986 - val_accuracy: 0.8060
Epoch 169/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3830 - accuracy: 0.8207 - val_loss: 0.4153 - val_accuracy: 0.7967
Epoch 170/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3835 - accuracy: 0.8222 - val_loss: 0.4008 - val_accuracy: 0.8060
Epoch 171/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3823 - accuracy: 0.8222 - val_loss: 0.4058 - val_accuracy: 0.7991
Epoch 172/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3838 - accuracy: 0.8226 - val_loss: 0.4121 - val_accuracy: 0.8022
Epoch 173/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3818 - accuracy: 0.8243 - val_loss: 0.4097 - val_accuracy: 0.8068
Epoch 174/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3852 - accuracy: 0.8227 - val_loss: 0.3965 - val_accuracy: 0.8068
Epoch 175/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3852 - accuracy: 0.8192 - val_loss: 0.4027 - val_accuracy: 0.8057
Epoch 176/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3891 - accuracy: 0.8148 - val_loss: 0.4072 - val_accuracy: 0.8064
Epoch 177/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3915 - accuracy: 0.8118 - val_loss: 0.4297 - val_accuracy: 0.7890
Epoch 178/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3824 - accuracy: 0.8256 - val_loss: 0.4325 - val_accuracy: 0.7878
Epoch 179/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3863 - accuracy: 0.8204 - val_loss: 0.3980 - val_accuracy: 0.8064
Epoch 180/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3824 - accuracy: 0.8237 - val_loss: 0.4081 - val_accuracy: 0.7991
Epoch 181/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3861 - accuracy: 0.8189 - val_loss: 0.4207 - val_accuracy: 0.7971
Epoch 182/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3885 - accuracy: 0.8221 - val_loss: 0.4134 - val_accuracy: 0.8072
Epoch 183/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3844 - accuracy: 0.8211 - val_loss: 0.4032 - val_accuracy: 0.8060
Epoch 184/250
162/162 [==============================] - 4s 28ms/step - loss: 0.3841 - accuracy: 0.8255 - val_loss: 0.3982 - val_accuracy: 0.8064
Epoch 185/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3817 - accuracy: 0.8222 - val_loss: 0.4164 - val_accuracy: 0.7960
Epoch 186/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3854 - accuracy: 0.8213 - val_loss: 0.3954 - val_accuracy: 0.8076
Epoch 187/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3826 - accuracy: 0.8232 - val_loss: 0.4020 - val_accuracy: 0.8041
Epoch 188/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3841 - accuracy: 0.8205 - val_loss: 0.3970 - val_accuracy: 0.8053
Epoch 189/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3836 - accuracy: 0.8223 - val_loss: 0.4042 - val_accuracy: 0.8033
Epoch 190/250
162/162 [==============================] - 4s 27ms/step - loss: 0.3861 - accuracy: 0.8182 - val_loss: 0.3988 - val_accuracy: 0.8080
Epoch 191/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3811 - accuracy: 0.8209 - val_loss: 0.3974 - val_accuracy: 0.8041
Epoch 192/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3797 - accuracy: 0.8238 - val_loss: 0.4118 - val_accuracy: 0.8095
Epoch 193/250
162/162 [==============================] - 4s 28ms/step - loss: 0.3835 - accuracy: 0.8222 - val_loss: 0.4008 - val_accuracy: 0.8091
Epoch 194/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3851 - accuracy: 0.8218 - val_loss: 0.4025 - val_accuracy: 0.8080
Epoch 195/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3845 - accuracy: 0.8221 - val_loss: 0.4198 - val_accuracy: 0.7995
Epoch 196/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3827 - accuracy: 0.8229 - val_loss: 0.4021 - val_accuracy: 0.8076
Epoch 197/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3818 - accuracy: 0.8230 - val_loss: 0.4363 - val_accuracy: 0.7902
Epoch 198/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3839 - accuracy: 0.8211 - val_loss: 0.3978 - val_accuracy: 0.8068
Epoch 199/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3878 - accuracy: 0.8180 - val_loss: 0.3945 - val_accuracy: 0.8080
Epoch 200/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3846 - accuracy: 0.8223 - val_loss: 0.4157 - val_accuracy: 0.7967
Epoch 201/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3831 - accuracy: 0.8241 - val_loss: 0.4167 - val_accuracy: 0.7944
Epoch 202/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3867 - accuracy: 0.8209 - val_loss: 0.4100 - val_accuracy: 0.8060
Epoch 203/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3825 - accuracy: 0.8228 - val_loss: 0.3985 - val_accuracy: 0.8057
Epoch 204/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3826 - accuracy: 0.8222 - val_loss: 0.4265 - val_accuracy: 0.7855
Epoch 205/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3837 - accuracy: 0.8178 - val_loss: 0.4097 - val_accuracy: 0.8084
Epoch 206/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3831 - accuracy: 0.8217 - val_loss: 0.4020 - val_accuracy: 0.8060
Epoch 207/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3820 - accuracy: 0.8228 - val_loss: 0.4007 - val_accuracy: 0.8064
Epoch 208/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3853 - accuracy: 0.8186 - val_loss: 0.4132 - val_accuracy: 0.7998
Epoch 209/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3882 - accuracy: 0.8192 - val_loss: 0.4197 - val_accuracy: 0.8006
Epoch 210/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3807 - accuracy: 0.8249 - val_loss: 0.3972 - val_accuracy: 0.8084
Epoch 211/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3823 - accuracy: 0.8238 - val_loss: 0.4036 - val_accuracy: 0.8037
Epoch 212/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3846 - accuracy: 0.8208 - val_loss: 0.3985 - val_accuracy: 0.8111
Epoch 213/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3799 - accuracy: 0.8239 - val_loss: 0.4134 - val_accuracy: 0.8010
Epoch 214/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3808 - accuracy: 0.8238 - val_loss: 0.4032 - val_accuracy: 0.8099
Epoch 215/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3854 - accuracy: 0.8230 - val_loss: 0.3933 - val_accuracy: 0.8107
Epoch 216/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3791 - accuracy: 0.8248 - val_loss: 0.4062 - val_accuracy: 0.8006
Epoch 217/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3800 - accuracy: 0.8265 - val_loss: 0.4002 - val_accuracy: 0.8111
Epoch 218/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3854 - accuracy: 0.8228 - val_loss: 0.3929 - val_accuracy: 0.8076
Epoch 219/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3778 - accuracy: 0.8262 - val_loss: 0.4278 - val_accuracy: 0.7948
Epoch 220/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3849 - accuracy: 0.8207 - val_loss: 0.4273 - val_accuracy: 0.7882
Epoch 221/250
162/162 [==============================] - 5s 28ms/step - loss: 0.3813 - accuracy: 0.8220 - val_loss: 0.4009 - val_accuracy: 0.8064
Epoch 222/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3822 - accuracy: 0.8258 - val_loss: 0.3988 - val_accuracy: 0.8080
Epoch 223/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3793 - accuracy: 0.8269 - val_loss: 0.3960 - val_accuracy: 0.8099
Epoch 224/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3816 - accuracy: 0.8229 - val_loss: 0.4193 - val_accuracy: 0.8010
Epoch 225/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3823 - accuracy: 0.8233 - val_loss: 0.4032 - val_accuracy: 0.8087
Epoch 226/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3817 - accuracy: 0.8251 - val_loss: 0.4094 - val_accuracy: 0.8018
Epoch 227/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3804 - accuracy: 0.8244 - val_loss: 0.3976 - val_accuracy: 0.8037
Epoch 228/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3815 - accuracy: 0.8268 - val_loss: 0.3958 - val_accuracy: 0.8095
Epoch 229/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3829 - accuracy: 0.8250 - val_loss: 0.3995 - val_accuracy: 0.8146
Epoch 230/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3785 - accuracy: 0.8255 - val_loss: 0.3983 - val_accuracy: 0.8118
Epoch 231/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3799 - accuracy: 0.8250 - val_loss: 0.3979 - val_accuracy: 0.8103
Epoch 232/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3800 - accuracy: 0.8238 - val_loss: 0.4096 - val_accuracy: 0.8022
Epoch 233/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3765 - accuracy: 0.8273 - val_loss: 0.3927 - val_accuracy: 0.8130
Epoch 234/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3806 - accuracy: 0.8273 - val_loss: 0.3957 - val_accuracy: 0.8118
Epoch 235/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3788 - accuracy: 0.8289 - val_loss: 0.4061 - val_accuracy: 0.8099
Epoch 236/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3769 - accuracy: 0.8262 - val_loss: 0.3916 - val_accuracy: 0.8138
Epoch 237/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3805 - accuracy: 0.8258 - val_loss: 0.3986 - val_accuracy: 0.8107
Epoch 238/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3771 - accuracy: 0.8287 - val_loss: 0.4242 - val_accuracy: 0.7991
Epoch 239/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3804 - accuracy: 0.8247 - val_loss: 0.3964 - val_accuracy: 0.8103
Epoch 240/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3775 - accuracy: 0.8276 - val_loss: 0.3974 - val_accuracy: 0.8118
Epoch 241/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3766 - accuracy: 0.8270 - val_loss: 0.4111 - val_accuracy: 0.8010
Epoch 242/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3782 - accuracy: 0.8274 - val_loss: 0.3921 - val_accuracy: 0.8180
Epoch 243/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3806 - accuracy: 0.8278 - val_loss: 0.3927 - val_accuracy: 0.8177
Epoch 244/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3778 - accuracy: 0.8289 - val_loss: 0.3935 - val_accuracy: 0.8146
Epoch 245/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3787 - accuracy: 0.8268 - val_loss: 0.3965 - val_accuracy: 0.8157
Epoch 246/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3780 - accuracy: 0.8288 - val_loss: 0.3958 - val_accuracy: 0.8149
Epoch 247/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3828 - accuracy: 0.8209 - val_loss: 0.3970 - val_accuracy: 0.8068
Epoch 248/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3760 - accuracy: 0.8273 - val_loss: 0.3926 - val_accuracy: 0.8134
Epoch 249/250
162/162 [==============================] - 4s 26ms/step - loss: 0.3841 - accuracy: 0.8225 - val_loss: 0.3933 - val_accuracy: 0.8111
Epoch 250/250
162/162 [==============================] - 4s 27ms/step - loss: 0.3765 - accuracy: 0.8291 - val_loss: 0.4005 - val_accuracy: 0.8146

Step 7.1.2: Evaluation

plt.style.use('fivethirtyeight')
epochs = [i for i in range(250)]
fig , ax = plt.subplots(1,2)
train_acc = history_2_g.history['accuracy']
train_loss = history_2_g.history['loss']
test_acc = history_2_g.history['val_accuracy']
test_loss = history_2_g.history['val_loss']

fig.set_size_inches(20,6)
ax[0].plot(epochs , train_loss , label = 'Training Loss')
ax[0].plot(epochs , test_loss , label = 'Testing Loss')
ax[0].set_title('Training & Testing Loss')
ax[0].legend()
ax[0].set_xlabel("Epochs")

ax[1].plot(epochs , train_acc , label = 'Training Accuracy')
ax[1].plot(epochs , test_acc , label = 'Testing Accuracy')
ax[1].set_title('Training & Testing Accuracy')
ax[1].legend()
ax[1].set_xlabel("Epochs")
plt.show()

Step 7.1.3: Save Model

# Save Model

lstm_1_g.save('Trained Models/lstm_1_g.h5')

Step 7.2: For Emotion Classification

x_train, x_test, y_train, y_test = X_train, X_test, y2_train, y2_test

size = 13
x_train = np.array(x_train)
y_train = np.array(y_train)
x_test = np.array(x_test) 
y_test = np.array(y_test)
x_train = x_train.reshape(x_train.shape[0], size, 1)
x_test = x_test.reshape(x_test.shape[0], size, 1)
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)

# Create Model

lstm_1_e = Sequential()
lstm_1_e.add(Dense(256,  activation='relu', input_shape=(x_train.shape[1],1)))
lstm_1_e.add(Dense(128, activation='relu'))
lstm_1_e.add(Dense(64, activation='relu'))
lstm_1_e.add(Dense(32, activation='relu'))
lstm_1_e.add(Flatten())
lstm_1_e.add(Dense(16, activation='relu'))
lstm_1_e.add(Dense(10, activation='relu'))
lstm_1_e.add(Dense(8, activation='relu'))
lstm_1_e.add(Flatten())
lstm_1_e.add(Dense(7, activation='softmax'))
lstm_1_e.compile(optimizer = 'adam' , loss = 'categorical_crossentropy' , metrics = ['accuracy'])
lstm_1_e.summary()

Model: "sequential_3"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_11 (Dense)             (None, 13, 256)           512       
_________________________________________________________________
dense_12 (Dense)             (None, 13, 128)           32896     
_________________________________________________________________
dense_13 (Dense)             (None, 13, 64)            8256      
_________________________________________________________________
dense_14 (Dense)             (None, 13, 32)            2080      
_________________________________________________________________
flatten_3 (Flatten)          (None, 416)               0         
_________________________________________________________________
dense_15 (Dense)             (None, 16)                6672      
_________________________________________________________________
dense_16 (Dense)             (None, 10)                170       
_________________________________________________________________
dense_17 (Dense)             (None, 8)                 88        
_________________________________________________________________
flatten_4 (Flatten)          (None, 8)                 0         
_________________________________________________________________
dense_18 (Dense)             (None, 7)                 63        
=================================================================
Total params: 50,737
Trainable params: 50,737
Non-trainable params: 0
_________________________________________________________________

Step 7.2.1: Training

history_2_e = lstm_1_e.fit(x_train, y_train, epochs=250, batch_size=16, validation_split=0.1)

Epoch 1/250
727/727 [==============================] - 15s 21ms/step - loss: 1.7595 - accuracy: 0.2800 - val_loss: 1.5582 - val_accuracy: 0.3901
Epoch 2/250
727/727 [==============================] - 15s 20ms/step - loss: 1.5375 - accuracy: 0.3900 - val_loss: 1.4874 - val_accuracy: 0.4319
Epoch 3/250
727/727 [==============================] - 15s 20ms/step - loss: 1.4745 - accuracy: 0.4134 - val_loss: 1.4613 - val_accuracy: 0.4404
Epoch 4/250
727/727 [==============================] - 15s 20ms/step - loss: 1.4336 - accuracy: 0.4366 - val_loss: 1.4151 - val_accuracy: 0.4567
Epoch 5/250
727/727 [==============================] - 15s 20ms/step - loss: 1.4032 - accuracy: 0.4563 - val_loss: 1.3890 - val_accuracy: 0.4582
Epoch 6/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3861 - accuracy: 0.4542 - val_loss: 1.3842 - val_accuracy: 0.4582
Epoch 7/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3706 - accuracy: 0.4667 - val_loss: 1.3717 - val_accuracy: 0.4551
Epoch 8/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3593 - accuracy: 0.4699 - val_loss: 1.3458 - val_accuracy: 0.4791
Epoch 9/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3477 - accuracy: 0.4699 - val_loss: 1.3408 - val_accuracy: 0.4752
Epoch 10/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3435 - accuracy: 0.4762 - val_loss: 1.3295 - val_accuracy: 0.4837
Epoch 11/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3329 - accuracy: 0.4804 - val_loss: 1.3288 - val_accuracy: 0.4923
Epoch 12/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3271 - accuracy: 0.4796 - val_loss: 1.3439 - val_accuracy: 0.4837
Epoch 13/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3202 - accuracy: 0.4829 - val_loss: 1.3470 - val_accuracy: 0.4706
Epoch 14/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3144 - accuracy: 0.4862 - val_loss: 1.3113 - val_accuracy: 0.4899
Epoch 15/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3120 - accuracy: 0.4865 - val_loss: 1.3354 - val_accuracy: 0.4551
Epoch 16/250
727/727 [==============================] - 15s 20ms/step - loss: 1.3044 - accuracy: 0.4920 - val_loss: 1.3280 - val_accuracy: 0.4745
Epoch 17/250
727/727 [==============================] - 15s 21ms/step - loss: 1.3023 - accuracy: 0.4912 - val_loss: 1.3002 - val_accuracy: 0.4837
Epoch 18/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2976 - accuracy: 0.4935 - val_loss: 1.3259 - val_accuracy: 0.4876
Epoch 19/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2920 - accuracy: 0.4920 - val_loss: 1.3167 - val_accuracy: 0.4837
Epoch 20/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2902 - accuracy: 0.4952 - val_loss: 1.3234 - val_accuracy: 0.4783
Epoch 21/250
727/727 [==============================] - 14s 20ms/step - loss: 1.2860 - accuracy: 0.4900 - val_loss: 1.2824 - val_accuracy: 0.4946
Epoch 22/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2852 - accuracy: 0.4940 - val_loss: 1.2751 - val_accuracy: 0.5062
Epoch 23/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2844 - accuracy: 0.4984 - val_loss: 1.3042 - val_accuracy: 0.4876
Epoch 24/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2787 - accuracy: 0.4961 - val_loss: 1.2778 - val_accuracy: 0.5031
Epoch 25/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2749 - accuracy: 0.4982 - val_loss: 1.2815 - val_accuracy: 0.4892
Epoch 26/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2712 - accuracy: 0.5011 - val_loss: 1.3010 - val_accuracy: 0.4907
Epoch 27/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2734 - accuracy: 0.5011 - val_loss: 1.3209 - val_accuracy: 0.4814
Epoch 28/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2656 - accuracy: 0.5031 - val_loss: 1.2789 - val_accuracy: 0.5046
Epoch 29/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2617 - accuracy: 0.5013 - val_loss: 1.3818 - val_accuracy: 0.4474
Epoch 30/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2626 - accuracy: 0.5014 - val_loss: 1.2768 - val_accuracy: 0.5062
Epoch 31/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2610 - accuracy: 0.5039 - val_loss: 1.2890 - val_accuracy: 0.4837
Epoch 32/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2627 - accuracy: 0.5059 - val_loss: 1.2690 - val_accuracy: 0.5039
Epoch 33/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2628 - accuracy: 0.5055 - val_loss: 1.2770 - val_accuracy: 0.5039
Epoch 34/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2536 - accuracy: 0.5051 - val_loss: 1.2776 - val_accuracy: 0.4930
Epoch 35/250
727/727 [==============================] - 14s 20ms/step - loss: 1.2579 - accuracy: 0.5054 - val_loss: 1.2688 - val_accuracy: 0.4992
Epoch 36/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2515 - accuracy: 0.5091 - val_loss: 1.2666 - val_accuracy: 0.5046
Epoch 37/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2512 - accuracy: 0.5084 - val_loss: 1.2824 - val_accuracy: 0.4899
Epoch 38/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2478 - accuracy: 0.5085 - val_loss: 1.3862 - val_accuracy: 0.4543
Epoch 39/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2487 - accuracy: 0.5027 - val_loss: 1.2525 - val_accuracy: 0.5062
Epoch 40/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2433 - accuracy: 0.5112 - val_loss: 1.2647 - val_accuracy: 0.5077
Epoch 41/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2474 - accuracy: 0.5041 - val_loss: 1.2666 - val_accuracy: 0.5139
Epoch 42/250
727/727 [==============================] - 15s 21ms/step - loss: 1.2361 - accuracy: 0.5136 - val_loss: 1.2922 - val_accuracy: 0.4946
Epoch 43/250
727/727 [==============================] - 15s 21ms/step - loss: 1.2352 - accuracy: 0.5141 - val_loss: 1.2703 - val_accuracy: 0.4954
Epoch 44/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2392 - accuracy: 0.5086 - val_loss: 1.2621 - val_accuracy: 0.5015
Epoch 45/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2380 - accuracy: 0.5074 - val_loss: 1.2774 - val_accuracy: 0.5101
Epoch 46/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2311 - accuracy: 0.5165 - val_loss: 1.2667 - val_accuracy: 0.5062
Epoch 47/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2274 - accuracy: 0.5150 - val_loss: 1.2976 - val_accuracy: 0.4907
Epoch 48/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2317 - accuracy: 0.5118 - val_loss: 1.2615 - val_accuracy: 0.4946
Epoch 49/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2289 - accuracy: 0.5118 - val_loss: 1.2949 - val_accuracy: 0.5039
Epoch 50/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2271 - accuracy: 0.5157 - val_loss: 1.2591 - val_accuracy: 0.5101
Epoch 51/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2223 - accuracy: 0.5148 - val_loss: 1.2471 - val_accuracy: 0.5124
Epoch 52/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2248 - accuracy: 0.5185 - val_loss: 1.2592 - val_accuracy: 0.5062
Epoch 53/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2235 - accuracy: 0.5185 - val_loss: 1.2565 - val_accuracy: 0.5093
Epoch 54/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2202 - accuracy: 0.5162 - val_loss: 1.2493 - val_accuracy: 0.5046
Epoch 55/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2203 - accuracy: 0.5222 - val_loss: 1.2804 - val_accuracy: 0.4977
Epoch 56/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2158 - accuracy: 0.5236 - val_loss: 1.2874 - val_accuracy: 0.5046
Epoch 57/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2231 - accuracy: 0.5178 - val_loss: 1.3122 - val_accuracy: 0.4907
Epoch 58/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2156 - accuracy: 0.5245 - val_loss: 1.2426 - val_accuracy: 0.5263
Epoch 59/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2108 - accuracy: 0.5245 - val_loss: 1.2634 - val_accuracy: 0.5046
Epoch 60/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2130 - accuracy: 0.5223 - val_loss: 1.2512 - val_accuracy: 0.5116
Epoch 61/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2106 - accuracy: 0.5231 - val_loss: 1.2433 - val_accuracy: 0.5093
Epoch 62/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2062 - accuracy: 0.5251 - val_loss: 1.2499 - val_accuracy: 0.5046
Epoch 63/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2075 - accuracy: 0.5221 - val_loss: 1.2402 - val_accuracy: 0.5054
Epoch 64/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2093 - accuracy: 0.5267 - val_loss: 1.2571 - val_accuracy: 0.5132
Epoch 65/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2013 - accuracy: 0.5271 - val_loss: 1.2359 - val_accuracy: 0.5248
Epoch 66/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2077 - accuracy: 0.5192 - val_loss: 1.2316 - val_accuracy: 0.5317
Epoch 67/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2016 - accuracy: 0.5252 - val_loss: 1.2416 - val_accuracy: 0.5163
Epoch 68/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2049 - accuracy: 0.5236 - val_loss: 1.2545 - val_accuracy: 0.5077
Epoch 69/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2031 - accuracy: 0.5253 - val_loss: 1.2876 - val_accuracy: 0.5000
Epoch 70/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2064 - accuracy: 0.5244 - val_loss: 1.2481 - val_accuracy: 0.5000
Epoch 71/250
727/727 [==============================] - 15s 20ms/step - loss: 1.2031 - accuracy: 0.5268 - val_loss: 1.2847 - val_accuracy: 0.5015
Epoch 72/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1997 - accuracy: 0.5274 - val_loss: 1.2214 - val_accuracy: 0.5248
Epoch 73/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1971 - accuracy: 0.5297 - val_loss: 1.2378 - val_accuracy: 0.5062
Epoch 74/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1969 - accuracy: 0.5265 - val_loss: 1.2786 - val_accuracy: 0.4977
Epoch 75/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1947 - accuracy: 0.5281 - val_loss: 1.2371 - val_accuracy: 0.5085
Epoch 76/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1978 - accuracy: 0.5247 - val_loss: 1.2349 - val_accuracy: 0.5186
Epoch 77/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1948 - accuracy: 0.5275 - val_loss: 1.2326 - val_accuracy: 0.5232
Epoch 78/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1891 - accuracy: 0.5271 - val_loss: 1.2196 - val_accuracy: 0.5341
Epoch 79/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1933 - accuracy: 0.5281 - val_loss: 1.2548 - val_accuracy: 0.5139
Epoch 80/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1915 - accuracy: 0.5327 - val_loss: 1.2581 - val_accuracy: 0.5108
Epoch 81/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1925 - accuracy: 0.5244 - val_loss: 1.2259 - val_accuracy: 0.5286
Epoch 82/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1925 - accuracy: 0.5323 - val_loss: 1.2209 - val_accuracy: 0.5248
Epoch 83/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1910 - accuracy: 0.5304 - val_loss: 1.2266 - val_accuracy: 0.5163
Epoch 84/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1916 - accuracy: 0.5290 - val_loss: 1.2499 - val_accuracy: 0.5031
Epoch 85/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1867 - accuracy: 0.5324 - val_loss: 1.2621 - val_accuracy: 0.5124
Epoch 86/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1895 - accuracy: 0.5338 - val_loss: 1.2300 - val_accuracy: 0.5209
Epoch 87/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1884 - accuracy: 0.5330 - val_loss: 1.2287 - val_accuracy: 0.5224
Epoch 88/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1828 - accuracy: 0.5330 - val_loss: 1.2186 - val_accuracy: 0.5139
Epoch 89/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1848 - accuracy: 0.5308 - val_loss: 1.2289 - val_accuracy: 0.5286
Epoch 90/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1872 - accuracy: 0.5333 - val_loss: 1.2189 - val_accuracy: 0.5294
Epoch 91/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1802 - accuracy: 0.5364 - val_loss: 1.2472 - val_accuracy: 0.5193
Epoch 92/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1794 - accuracy: 0.5380 - val_loss: 1.2142 - val_accuracy: 0.5193
Epoch 93/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1829 - accuracy: 0.5321 - val_loss: 1.2246 - val_accuracy: 0.5410
Epoch 94/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1834 - accuracy: 0.5338 - val_loss: 1.2300 - val_accuracy: 0.5186
Epoch 95/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1803 - accuracy: 0.5344 - val_loss: 1.2656 - val_accuracy: 0.5062
Epoch 96/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1806 - accuracy: 0.5327 - val_loss: 1.2302 - val_accuracy: 0.5070
Epoch 97/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1797 - accuracy: 0.5349 - val_loss: 1.2364 - val_accuracy: 0.5108
Epoch 98/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1773 - accuracy: 0.5357 - val_loss: 1.2248 - val_accuracy: 0.5341
Epoch 99/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1744 - accuracy: 0.5344 - val_loss: 1.2123 - val_accuracy: 0.5170
Epoch 100/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1779 - accuracy: 0.5323 - val_loss: 1.2161 - val_accuracy: 0.5294
Epoch 101/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1804 - accuracy: 0.5348 - val_loss: 1.2228 - val_accuracy: 0.5356
Epoch 102/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1781 - accuracy: 0.5371 - val_loss: 1.2116 - val_accuracy: 0.5356
Epoch 103/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1730 - accuracy: 0.5378 - val_loss: 1.2328 - val_accuracy: 0.5178
Epoch 104/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1768 - accuracy: 0.5297 - val_loss: 1.2065 - val_accuracy: 0.5248
Epoch 105/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1725 - accuracy: 0.5356 - val_loss: 1.2052 - val_accuracy: 0.5379
Epoch 106/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1716 - accuracy: 0.5411 - val_loss: 1.2218 - val_accuracy: 0.5271
Epoch 107/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1750 - accuracy: 0.5392 - val_loss: 1.2093 - val_accuracy: 0.5193
Epoch 108/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1728 - accuracy: 0.5379 - val_loss: 1.2121 - val_accuracy: 0.5387
Epoch 109/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1745 - accuracy: 0.5379 - val_loss: 1.2394 - val_accuracy: 0.5279
Epoch 110/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1704 - accuracy: 0.5407 - val_loss: 1.2106 - val_accuracy: 0.5395
Epoch 111/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1699 - accuracy: 0.5396 - val_loss: 1.2024 - val_accuracy: 0.5356
Epoch 112/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1679 - accuracy: 0.5416 - val_loss: 1.2287 - val_accuracy: 0.5302
Epoch 113/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1744 - accuracy: 0.5380 - val_loss: 1.2775 - val_accuracy: 0.5178
Epoch 114/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1713 - accuracy: 0.5392 - val_loss: 1.2229 - val_accuracy: 0.5302
Epoch 115/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1711 - accuracy: 0.5390 - val_loss: 1.2470 - val_accuracy: 0.5224
Epoch 116/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1668 - accuracy: 0.5404 - val_loss: 1.2181 - val_accuracy: 0.5224
Epoch 117/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1720 - accuracy: 0.5404 - val_loss: 1.2085 - val_accuracy: 0.5387
Epoch 118/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1678 - accuracy: 0.5389 - val_loss: 1.2232 - val_accuracy: 0.5224
Epoch 119/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1658 - accuracy: 0.5400 - val_loss: 1.2409 - val_accuracy: 0.5209
Epoch 120/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1681 - accuracy: 0.5379 - val_loss: 1.2013 - val_accuracy: 0.5526
Epoch 121/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1665 - accuracy: 0.5443 - val_loss: 1.2249 - val_accuracy: 0.5255
Epoch 122/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1642 - accuracy: 0.5420 - val_loss: 1.2127 - val_accuracy: 0.5356
Epoch 123/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1627 - accuracy: 0.5447 - val_loss: 1.2069 - val_accuracy: 0.5441
Epoch 124/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1634 - accuracy: 0.5417 - val_loss: 1.2195 - val_accuracy: 0.5193
Epoch 125/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1627 - accuracy: 0.5436 - val_loss: 1.2088 - val_accuracy: 0.5433
Epoch 126/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1612 - accuracy: 0.5445 - val_loss: 1.2112 - val_accuracy: 0.5410
Epoch 127/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1653 - accuracy: 0.5421 - val_loss: 1.2064 - val_accuracy: 0.5433
Epoch 128/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1568 - accuracy: 0.5481 - val_loss: 1.2061 - val_accuracy: 0.5286
Epoch 129/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1620 - accuracy: 0.5418 - val_loss: 1.2216 - val_accuracy: 0.5271
Epoch 130/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1617 - accuracy: 0.5389 - val_loss: 1.2075 - val_accuracy: 0.5426
Epoch 131/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1593 - accuracy: 0.5451 - val_loss: 1.2218 - val_accuracy: 0.5263
Epoch 132/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1587 - accuracy: 0.5390 - val_loss: 1.2385 - val_accuracy: 0.5263
Epoch 133/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1582 - accuracy: 0.5414 - val_loss: 1.2226 - val_accuracy: 0.4961
Epoch 134/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1593 - accuracy: 0.5449 - val_loss: 1.2121 - val_accuracy: 0.5294
Epoch 135/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1552 - accuracy: 0.5438 - val_loss: 1.2397 - val_accuracy: 0.5441
Epoch 136/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1598 - accuracy: 0.5424 - val_loss: 1.2246 - val_accuracy: 0.5410
Epoch 137/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1561 - accuracy: 0.5419 - val_loss: 1.2240 - val_accuracy: 0.5286
Epoch 138/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1563 - accuracy: 0.5437 - val_loss: 1.2345 - val_accuracy: 0.5224
Epoch 139/250
727/727 [==============================] - 16s 22ms/step - loss: 1.1575 - accuracy: 0.5421 - val_loss: 1.2237 - val_accuracy: 0.5248
Epoch 140/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1541 - accuracy: 0.5452 - val_loss: 1.2345 - val_accuracy: 0.5341
Epoch 141/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1556 - accuracy: 0.5437 - val_loss: 1.2050 - val_accuracy: 0.5348
Epoch 142/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1516 - accuracy: 0.5435 - val_loss: 1.2448 - val_accuracy: 0.5286
Epoch 143/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1548 - accuracy: 0.5392 - val_loss: 1.2154 - val_accuracy: 0.5472
Epoch 144/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1550 - accuracy: 0.5472 - val_loss: 1.2646 - val_accuracy: 0.5070
Epoch 145/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1572 - accuracy: 0.5443 - val_loss: 1.2339 - val_accuracy: 0.5132
Epoch 146/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1543 - accuracy: 0.5459 - val_loss: 1.2382 - val_accuracy: 0.5310
Epoch 147/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1527 - accuracy: 0.5454 - val_loss: 1.2137 - val_accuracy: 0.5193
Epoch 148/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1499 - accuracy: 0.5469 - val_loss: 1.2334 - val_accuracy: 0.5294
Epoch 149/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1535 - accuracy: 0.5416 - val_loss: 1.2084 - val_accuracy: 0.5387
Epoch 150/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1514 - accuracy: 0.5438 - val_loss: 1.2190 - val_accuracy: 0.5271
Epoch 151/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1504 - accuracy: 0.5478 - val_loss: 1.2908 - val_accuracy: 0.5108
Epoch 152/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1574 - accuracy: 0.5419 - val_loss: 1.2162 - val_accuracy: 0.5294
Epoch 153/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1500 - accuracy: 0.5489 - val_loss: 1.2036 - val_accuracy: 0.5348
Epoch 154/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1502 - accuracy: 0.5486 - val_loss: 1.2018 - val_accuracy: 0.5387
Epoch 155/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1475 - accuracy: 0.5487 - val_loss: 1.3168 - val_accuracy: 0.4930
Epoch 156/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1523 - accuracy: 0.5471 - val_loss: 1.2090 - val_accuracy: 0.5271
Epoch 157/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1495 - accuracy: 0.5511 - val_loss: 1.2442 - val_accuracy: 0.5201
Epoch 158/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1517 - accuracy: 0.5461 - val_loss: 1.2190 - val_accuracy: 0.5248
Epoch 159/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1520 - accuracy: 0.5430 - val_loss: 1.2196 - val_accuracy: 0.5333
Epoch 160/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1441 - accuracy: 0.5500 - val_loss: 1.2220 - val_accuracy: 0.5178
Epoch 161/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1483 - accuracy: 0.5493 - val_loss: 1.2205 - val_accuracy: 0.5379
Epoch 162/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1453 - accuracy: 0.5482 - val_loss: 1.2472 - val_accuracy: 0.5255
Epoch 163/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1448 - accuracy: 0.5498 - val_loss: 1.2193 - val_accuracy: 0.5402
Epoch 164/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1521 - accuracy: 0.5482 - val_loss: 1.2399 - val_accuracy: 0.5101
Epoch 165/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1453 - accuracy: 0.5496 - val_loss: 1.2211 - val_accuracy: 0.5333
Epoch 166/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1477 - accuracy: 0.5476 - val_loss: 1.2201 - val_accuracy: 0.5286
Epoch 167/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1442 - accuracy: 0.5504 - val_loss: 1.2386 - val_accuracy: 0.5132
Epoch 168/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1463 - accuracy: 0.5489 - val_loss: 1.2173 - val_accuracy: 0.5395
Epoch 169/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1443 - accuracy: 0.5447 - val_loss: 1.2316 - val_accuracy: 0.5255
Epoch 170/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1485 - accuracy: 0.5486 - val_loss: 1.2120 - val_accuracy: 0.5372
Epoch 171/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1428 - accuracy: 0.5499 - val_loss: 1.2321 - val_accuracy: 0.5263
Epoch 172/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1424 - accuracy: 0.5471 - val_loss: 1.2187 - val_accuracy: 0.5356
Epoch 173/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1384 - accuracy: 0.5504 - val_loss: 1.2156 - val_accuracy: 0.5325
Epoch 174/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1437 - accuracy: 0.5508 - val_loss: 1.2725 - val_accuracy: 0.5155
Epoch 175/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1480 - accuracy: 0.5493 - val_loss: 1.2179 - val_accuracy: 0.5240
Epoch 176/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1403 - accuracy: 0.5535 - val_loss: 1.2240 - val_accuracy: 0.5232
Epoch 177/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1440 - accuracy: 0.5509 - val_loss: 1.2189 - val_accuracy: 0.5271
Epoch 178/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1386 - accuracy: 0.5518 - val_loss: 1.2515 - val_accuracy: 0.5163
Epoch 179/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1359 - accuracy: 0.5543 - val_loss: 1.2071 - val_accuracy: 0.5217
Epoch 180/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1380 - accuracy: 0.5526 - val_loss: 1.2599 - val_accuracy: 0.5132
Epoch 181/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1412 - accuracy: 0.5520 - val_loss: 1.2036 - val_accuracy: 0.5418
Epoch 182/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1421 - accuracy: 0.5524 - val_loss: 1.2452 - val_accuracy: 0.5193
Epoch 183/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1412 - accuracy: 0.5486 - val_loss: 1.2148 - val_accuracy: 0.5480
Epoch 184/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1400 - accuracy: 0.5526 - val_loss: 1.2093 - val_accuracy: 0.5395
Epoch 185/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1415 - accuracy: 0.5513 - val_loss: 1.2193 - val_accuracy: 0.5294
Epoch 186/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1386 - accuracy: 0.5523 - val_loss: 1.2152 - val_accuracy: 0.5170
Epoch 187/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1362 - accuracy: 0.5595 - val_loss: 1.2962 - val_accuracy: 0.5031
Epoch 188/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1372 - accuracy: 0.5520 - val_loss: 1.2514 - val_accuracy: 0.5170
Epoch 189/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1359 - accuracy: 0.5539 - val_loss: 1.2082 - val_accuracy: 0.5302
Epoch 190/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1332 - accuracy: 0.5557 - val_loss: 1.2051 - val_accuracy: 0.5410
Epoch 191/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1393 - accuracy: 0.5508 - val_loss: 1.2053 - val_accuracy: 0.5426
Epoch 192/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1362 - accuracy: 0.5542 - val_loss: 1.2161 - val_accuracy: 0.5503
Epoch 193/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1377 - accuracy: 0.5489 - val_loss: 1.2533 - val_accuracy: 0.5054
Epoch 194/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1381 - accuracy: 0.5523 - val_loss: 1.2495 - val_accuracy: 0.5085
Epoch 195/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1326 - accuracy: 0.5558 - val_loss: 1.2216 - val_accuracy: 0.5294
Epoch 196/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1374 - accuracy: 0.5540 - val_loss: 1.2161 - val_accuracy: 0.5255
Epoch 197/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1311 - accuracy: 0.5547 - val_loss: 1.2249 - val_accuracy: 0.5232
Epoch 198/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1327 - accuracy: 0.5547 - val_loss: 1.2556 - val_accuracy: 0.5209
Epoch 199/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1340 - accuracy: 0.5552 - val_loss: 1.2166 - val_accuracy: 0.5155
Epoch 200/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1386 - accuracy: 0.5547 - val_loss: 1.2131 - val_accuracy: 0.5163
Epoch 201/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1336 - accuracy: 0.5559 - val_loss: 1.2108 - val_accuracy: 0.5333
Epoch 202/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1303 - accuracy: 0.5561 - val_loss: 1.2122 - val_accuracy: 0.5279
Epoch 203/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1341 - accuracy: 0.5547 - val_loss: 1.2028 - val_accuracy: 0.5325
Epoch 204/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1307 - accuracy: 0.5574 - val_loss: 1.2200 - val_accuracy: 0.5310
Epoch 205/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1326 - accuracy: 0.5582 - val_loss: 1.2190 - val_accuracy: 0.5372
Epoch 206/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1346 - accuracy: 0.5561 - val_loss: 1.2343 - val_accuracy: 0.5348
Epoch 207/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1321 - accuracy: 0.5548 - val_loss: 1.2212 - val_accuracy: 0.5356
Epoch 208/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1381 - accuracy: 0.5548 - val_loss: 1.2033 - val_accuracy: 0.5279
Epoch 209/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1338 - accuracy: 0.5524 - val_loss: 1.2123 - val_accuracy: 0.5232
Epoch 210/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1354 - accuracy: 0.5545 - val_loss: 1.2043 - val_accuracy: 0.5325
Epoch 211/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1332 - accuracy: 0.5507 - val_loss: 1.2262 - val_accuracy: 0.5186
Epoch 212/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1316 - accuracy: 0.5563 - val_loss: 1.2086 - val_accuracy: 0.5224
Epoch 213/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1306 - accuracy: 0.5563 - val_loss: 1.2030 - val_accuracy: 0.5348
Epoch 214/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1351 - accuracy: 0.5538 - val_loss: 1.2123 - val_accuracy: 0.5294
Epoch 215/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1291 - accuracy: 0.5541 - val_loss: 1.2170 - val_accuracy: 0.5294
Epoch 216/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1268 - accuracy: 0.5590 - val_loss: 1.2104 - val_accuracy: 0.5271
Epoch 217/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1339 - accuracy: 0.5535 - val_loss: 1.1993 - val_accuracy: 0.5294
Epoch 218/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1311 - accuracy: 0.5573 - val_loss: 1.2121 - val_accuracy: 0.5271
Epoch 219/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1251 - accuracy: 0.5544 - val_loss: 1.2256 - val_accuracy: 0.5255
Epoch 220/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1312 - accuracy: 0.5547 - val_loss: 1.2118 - val_accuracy: 0.5387
Epoch 221/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1267 - accuracy: 0.5585 - val_loss: 1.2262 - val_accuracy: 0.5333
Epoch 222/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1307 - accuracy: 0.5573 - val_loss: 1.2224 - val_accuracy: 0.5348
Epoch 223/250
727/727 [==============================] - 14s 20ms/step - loss: 1.1294 - accuracy: 0.5552 - val_loss: 1.2062 - val_accuracy: 0.5387
Epoch 224/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1266 - accuracy: 0.5557 - val_loss: 1.2408 - val_accuracy: 0.5217
Epoch 225/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1322 - accuracy: 0.5548 - val_loss: 1.2251 - val_accuracy: 0.5209
Epoch 226/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1299 - accuracy: 0.5574 - val_loss: 1.2057 - val_accuracy: 0.5255
Epoch 227/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1294 - accuracy: 0.5574 - val_loss: 1.2090 - val_accuracy: 0.5279
Epoch 228/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1271 - accuracy: 0.5547 - val_loss: 1.2445 - val_accuracy: 0.5116
Epoch 229/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1298 - accuracy: 0.5577 - val_loss: 1.2145 - val_accuracy: 0.5402
Epoch 230/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1263 - accuracy: 0.5618 - val_loss: 1.1936 - val_accuracy: 0.5379
Epoch 231/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1260 - accuracy: 0.5578 - val_loss: 1.2305 - val_accuracy: 0.5201
Epoch 232/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1260 - accuracy: 0.5585 - val_loss: 1.2152 - val_accuracy: 0.5163
Epoch 233/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1276 - accuracy: 0.5589 - val_loss: 1.2182 - val_accuracy: 0.5302
Epoch 234/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1282 - accuracy: 0.5557 - val_loss: 1.2155 - val_accuracy: 0.5379
Epoch 235/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1244 - accuracy: 0.5593 - val_loss: 1.2006 - val_accuracy: 0.5325
Epoch 236/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1261 - accuracy: 0.5586 - val_loss: 1.2081 - val_accuracy: 0.5255
Epoch 237/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1251 - accuracy: 0.5581 - val_loss: 1.2025 - val_accuracy: 0.5441
Epoch 238/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1342 - accuracy: 0.5550 - val_loss: 1.2158 - val_accuracy: 0.5232
Epoch 239/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1280 - accuracy: 0.5537 - val_loss: 1.2302 - val_accuracy: 0.5248
Epoch 240/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1215 - accuracy: 0.5615 - val_loss: 1.2061 - val_accuracy: 0.5310
Epoch 241/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1220 - accuracy: 0.5583 - val_loss: 1.1986 - val_accuracy: 0.5433
Epoch 242/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1231 - accuracy: 0.5582 - val_loss: 1.2106 - val_accuracy: 0.5418
Epoch 243/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1277 - accuracy: 0.5561 - val_loss: 1.2089 - val_accuracy: 0.5325
Epoch 244/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1227 - accuracy: 0.5607 - val_loss: 1.2190 - val_accuracy: 0.5279
Epoch 245/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1239 - accuracy: 0.5579 - val_loss: 1.2027 - val_accuracy: 0.5325
Epoch 246/250
727/727 [==============================] - 15s 21ms/step - loss: 1.1215 - accuracy: 0.5628 - val_loss: 1.2047 - val_accuracy: 0.5317
Epoch 247/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1261 - accuracy: 0.5576 - val_loss: 1.3110 - val_accuracy: 0.5046
Epoch 248/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1193 - accuracy: 0.5622 - val_loss: 1.2234 - val_accuracy: 0.5232
Epoch 249/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1255 - accuracy: 0.5588 - val_loss: 1.2266 - val_accuracy: 0.5224
Epoch 250/250
727/727 [==============================] - 15s 20ms/step - loss: 1.1218 - accuracy: 0.5614 - val_loss: 1.2101 - val_accuracy: 0.5410

Step 7.2.2: Evalution

plt.style.use('fivethirtyeight')
epochs = [i for i in range(250)]
fig , ax = plt.subplots(1,2)
train_acc = history_2_e.history['accuracy']
train_loss = history_2_e.history['loss']
test_acc = history_2_e.history['val_accuracy']
test_loss = history_2_e.history['val_loss']

fig.set_size_inches(20,6)
ax[0].plot(epochs , train_loss , label = 'Training Loss')
ax[0].plot(epochs , test_loss , label = 'Testing Loss')
ax[0].set_title('Training & Testing Loss')
ax[0].legend()
ax[0].set_xlabel("Epochs")

ax[1].plot(epochs , train_acc , label = 'Training Accuracy')
ax[1].plot(epochs , test_acc , label = 'Testing Accuracy')
ax[1].set_title('Training & Testing Accuracy')
ax[1].legend()
ax[1].set_xlabel("Epochs")
plt.show()

Step 7.2.3: Save Model

# Save Model

lstm_1_g.save('Trained Models/lstm_1_e.h5')

Step 8: Comparision of CNN and LSTM

print("\nPerformance with Models")
print("=======================\n")
x=PrettyTable()
x.add_column("Classifier Name",["CNN (Gender)","CNN (Emotion)","LSTM (Gender)", "LSTM (Emotion)"])
x.add_column("Accuracy Score",[history_1_g.history['accuracy'][-1],history_1_e.history['accuracy'][-1],history_2_g.history['accuracy'][-1],history_2_e.history['accuracy'][-1]])
print(x)

x = ["CNN (Gender)","CNN (Emotion)","LSTM (Gender)", "LSTM (Emotion)"]
y = [round(history_1_g.history['accuracy'][-1],2),round(history_1_e.history['accuracy'][-1],2),round(history_2_g.history['accuracy'][-1],2),round(history_2_e.history['accuracy'][-1],2)]

plt.rcParams["figure.figsize"] = (14,5)
fig, ax = plt.subplots() 
width = 0.25
ind = np.arange(len(y))
ax.barh(ind, y, width, color="blue")
ax.set_yticks((ind+width/2)-0.1)
ax.set_yticklabels(x, minor=False)
for i, v in enumerate(y):
    ax.text(v + 0.01, i , str(v), color='blue', fontweight='bold')
plt.title('Comparision')
plt.xlabel('Accuracy Score')
plt.ylabel('Classifiers')      
plt.show()

Performance with Models
=======================

+-----------------+--------------------+
| Classifier Name |   Accuracy Score   |
+-----------------+--------------------+
|   CNN (Gender)  | 0.9777777791023254 |
|  CNN (Emotion)  | 0.7268292903900146 |
|  LSTM (Gender)  | 0.8290747404098511 |
|  LSTM (Emotion) | 0.5613868832588196 |
+-----------------+--------------------+